From shillcoc at mry.rbd.com  Sun Aug 29 02:46:33 1999
From: shillcoc at mry.rbd.com (Scott Shillcock)
Date: Sat, 28 Aug 1999 23:46:33 -0700
Subject: Need help with PyRun_SimpleFile
Message-ID: <37C8D749.15FB7C93@mry.rbd.com>

Hello,

Here is my very simple program:

#include <iostream.h>
#include "Python.h"

void
main (int argc, char **argv) {

   Py_Initialize ();
   FILE *fp = fopen ("test.py", "r+");
   cerr << "Before PyRun_SimpleFile" << endl;
   PyRun_SimpleFile (fp, "test.py");
   cerr << "After PyRun_SimpleFile" << endl;
   Py_Exit(0);
}

I can compile this into an exe file. I also created a very simple
test.py file that just prints one line of text. When I run my program it
crashes on the PyRun_SimpleFile() every time. I get the first cerr but
not the second.

What I am doing wrong? I have tried as many variations on this at I
could think of, all with the same result.

Thanks,
Scott

-- 
Scott Shillcock      shillcoc at rbd.com
Reality by Design         www.rbd.com



From webmaster at chatbase.com  Sun Aug 15 21:14:40 1999
From: webmaster at chatbase.com (Tim Greer)
Date: Sun, 15 Aug 1999 18:14:40 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7rb1nc.cf9.jstevens@bamboo.verinet.com>
Message-ID: <37B76600.FECF5FEB@chatbase.com>

John Stevens wrote:
> 
> On 14 Aug 1999 04:08:05 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
> >That could simply have been a reference. Or a symbolic reference.
> >
> >What is fundamental is that a @ tacked on the front indicates that it is an
> >array.
> 
> :-)
> 
> What is so amusing about that, is that you can say that with a straight
> face!
> 
> :-)
> 
> >So given @$fred, even with no knowledge of what that exactly means
> >you should be able to tell that it is somehow treating $fred as an array.
> 
> No, what any reasonable person would do would be to grab for his
> Perl book. . .

Until they learn the language enough to know these things. If you judge
a language based only on what you remember in the smallest amount of
learning time you've spent, then you're limiting yourself in too many
ways to bother mentioning. Some people have a better comprehension for
certain languages over others. It's really that simple. Some people are
scared off, or disillusioned too quickly. Simply put, they haven't spent
the proper amount of time (some languages do take longer to learn well
enough to not have to use reference material every few minutes), or they
weren't taught properly, or it's just something that they have less of a
comprehension for, or perhaps a lack of desire to really get so involved
when they could be using other languages. Sometimes that limits you and
your ability. You only limit yourself at that point, because you could
be writing better programs in a language that's probably better suited
for the task, a language that confuses you, because you don't understand
it as quickly as another. That's fine, and for some people, it's not
worth the time invested to pout so much time into learning for more
reasons that are equally as pointless to mention. Some people think
learning Assembly is a complete waste of time.. Why learn it when you
already know enough C? Right? Wrong. If you have time and reason or
desire, learn it. Even if you'll never use it, I'd suggest you learn it.
It might not be easy, but in most circumstances, you'll probably end up
writing better C programs... You'd have to learn Assembly to know what
I'm talking about... The only real drawback, is the fact that you'll
cringe when you realize what the C compiler is doing to your code. You
can't know this, because you either don't know the language, or know it
well enough to state such things. Sure, you find it more confusing then
Python, that's fine, that's you. However, in that case, you shouldn't
make statements that make you look foolish here, in a Perl NG, simply
based on the facts that you can't deal with the curly braces and such,
or the fact that _you_ personally can't remember some of the syntax and
whatnot. That's ridiculous and that's why this completely pointless
thread has continued for so long.

> >>Yes. . . is it a hash, or a scalar?  If it is a scalar, why
> >>is it called dict?  If it is a hash, then why is it prefixed
> >>by $?  If this is a reference instead of a scalar, then why
> >>doesn't it have it's own special prefix character.  ;->
> >
> >It's a scalar. It is named dict because TomC called it that.
> 
> Yes.  My point exactly.

Yes, well, tell me, _what_ is it to be called? You should know how much
Tom is involved in Perl and how much he contributes. It's not a bad idea
or a wrong idea to call it something _you_ personally don't agree with.
You are complaining about "lists" in Perl? It's what it's called in
_this_ language, _this_ is _not_ Python. Do you just use this as an
example, based on the fact that it's not what you're _used to seeing_?
Or what you think it would be better called? This is a pretty high-level
language, and you are sort of contradicting yourself by saying how
difficult the syntax is to remember for _you_ (however that's relevant
to the rest of the world?), yet you think it should be called a more
technical term that _most_ people wouldn't understand? Isn't that at
least _part_ of your point?
 
> >It is
> >also named that since it is a reference to a hash. I use code like this
> >in C quite a bit :
> 
> A reference to a hash. . . and yet TC claims that Perl is open to
> non-computer scientists.
> 
> :-)
> 
> Doesn't *ANYBODY* else see the irony in that?

How about the fact that you appear you be your own walking
contradiction? I believe that point is and was correct, and is to the
fact that; In Perl, people with no background in CS can learn enough,
right off, within a very short time, to be able to write some programs
that will accomplish some basic to intermediate tasks. That's not to say
that they will be able to know the more complex aspects of the language
with as much ease, and that goes without saying for _any_ language. Why
disagree with that?


> >If you know what it means then why do you continually get it wrong
> >throughout this thread?
> 
> I don't suppose that you realize that getting wrong simply
> proves (and illustrates) my point?

About you and your statements, yes. You aren't everyone, or the
majority. If enough people asked that some of the keywords and
definitions be changed and there was a real need for it, I'm sure it
would happen... I don't see that happening, because there's no reason,
because _most_ people here seem to understand it and remember the syntax
well enough to where they don't suffer from the same problem as you.
 
> I learned it.  I used it.  I haven't written a new Perl program
> in three months.
> 
> I come back to it, I get it wrong. . . do you see, yet,
> or do you just not get it?

I think the point is, is that you simply don't "get" Perl. You "get"
Python better, it's more comprehensible to you, and that's fine. Some
people find the opposite true. I myself, as both a Perl and Python
programmer (among other languages) look for and find the similarities.
There are more then a few if you really look at it. What's better for
what you're doing, is the question? What language do you find easier to
use personally? Why say one language is flawed over another based on
your own experience. Who did you learn from? Where did you learn from?
How much time did you spend? What, if any, do you have in prior
CS/programming backgrounds? Are you willing to dismiss some or many of
the things you were taught when you learned the prior, or do you feel
it's a better language based on the fact that you find it's closer to,
or easier to understand right off in comparison to the prior knowledge
you posses? Who are you to say?

<SNIP>

This is really stupid. Have you read any, many, or all of Tom's books
he's authored or co-authored? The man is not stupid and he knows what
he's talking about, which you have to admit is true when it comes to
Perl at least. Isn't it expected that he'll make points in defense of
Perl? You obviously don't like Perl as much as Python, and no one's
saying there's anything wrong with that.. The only flaw with your logic
in your posts, is the fact that it seems that you're speaking for others
and basing a lot of it on the fact, as you claim, that you yourself
personally find remembering the syntax difficult. Ok, so? That's you, I
can't think of anyone else that used Perl that has that problem, but I'm
sure more then a few do, just as with any other language. Why use Perl
when you've got Python? I don't know, maybe you'll enjoy Perl more, use
it more, or find it better suited for you or your task? Maybe your
particular program will run faster/better in perl rather then Python?
Maybe you can market it with more ease? Maybe something else? maybe your
program (for you, from you coding it) will run better coded in Python?
There's no point to this at all! If you want to compare it in that
manner, then tell me, _why_ bother with Perl OR Python, when you can use
C? Further, _why_ use C, when you can code it in Assembly? After all,
it's faster, you can make it do anything you want,
as-long-as-you-know-how. Just because you don't "know how", doesn't mean
it lacks something the language you _can_ code in better.

There was really no point to any of this. Making claims one way of the
other, about what you "believe" would be better named or worded does
little to no good. Note that my post in reply to you is/was completely
relevant and on-topic in regards to your post and this entire issue...
Note also, that I didn't have to use on piece of code to make this
point... after all, this is, to you as you've made it clear, to prove
your point, or disprove it. I'm just wondering _what_ your "point" was
supposed to be when it all comes down to it? I have been following the
thread, and I have yet to see a real point made.
-- 
Regards,
Tim Greer   : webmaster at chatbase.com  | software at linkworm.com
The ChatBase: http://www.chatbase.com | 250,000+ hits daily Worldwide!
TRG Software: http://www.linkworm.com | CGI scripts in Perl/C, & more.
Unix/NT/Novell Administration, Security, Web Design, ASP, SQL, & more.
Freelance Programming & Consulting, Musician, Martial Arts, +Sciences.



From JamesL at Lugoj.Com  Fri Aug 20 00:31:14 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Thu, 19 Aug 1999 21:31:14 -0700
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
Message-ID: <37BCDA12.6CB5B89A@Lugoj.Com>

Tom Christiansen wrote:
> There are no manpages for Python.

I hate HTML documentation. I want printed documentation or I get seriously
pissed too. Fortunately Python has a nice set of printable documents in a
variety of formats and paper sizes. See:

http://www.python.org/doc/current/download.html

You are fundamentally incorrect in stating that there is no man page for
Python. There is a man page for Python in my /usr/local/man/man1 directory
on my Linux box. It does not cover the language syntax or the huge library
reference, but then neither does the man page for c++ cover its syntax or
the STL. For your benefit I've included the Python man page as an
attachment. I hope this helps clear up any misunderstanding you may have
about Python documentation.
-------------- next part --------------
.TH PYTHON "7 September, 1997"
.SH NAME
python \- an interpreted, interactive, object-oriented programming language
.SH SYNOPSIS
.B python
[
.B \-d
]
[
.B \-i
]
[
.B \-O
]
[
.B \-S
]
[
.B \-u
]
[
.B \-v
]
[
.B \-X
]
.br
       [
.B \-c
.I command
|
.I script
|
\-
]
[
.I arguments
]
.SH DESCRIPTION
Python is an interpreted, interactive, object-oriented programming
language that combines remarkable power with very clear syntax.
For an introduction to programming in Python you are referred to the
Python Tutorial.
The Python Library Reference documents built-in and standard types,
constants, functions and modules.
Finally, the Python Reference Manual describes the syntax and
semantics of the core language in (perhaps too) much detail.
.PP
Python's basic power can be extended with your own modules written in
C or C++.
On most systems such modules may be dynamically loaded.
Python is also adaptable as an extension language for existing
applications.
See the internal documentation for hints.
.SH COMMAND LINE OPTIONS
.TP
.B \-d
Turn on parser debugging output (for wizards only, depending on
compilation options).
.TP
.B \-i
When a script is passed as first argument or the \fB\-c\fP option is
used, enter interactive mode after executing the script or the
command.  It does not read the $PYTHONSTARTUP file.  This can be
useful to inspect global variables or a stack trace when a script
raises an exception.
.TP
.B \-O
Turn on basic optimizations.  This changes the filename extension for
compiled (bytecode) files from
.I .pyc
to
.I pyo.
.TP
.B \-S
Disable the import of the module
.I site
and the site-dependent manipulations of
.I sys.path
that it entails.
.TP
.B \-u
Force stdin, stdout and stderr to be totally unbuffered.
.TP
.B \-v
Print a message each time a module is initialized, showing the place
(filename or built-in module) from which it is loaded.
.TP
.B \-X
Make the standard exceptions strings instead of classes.
Use for backward compatibility with old code only.
.TP
.BI "\-c " command
Specify the command to execute (see next section).
This terminates the option list (following options are passed as
arguments to the command).
.SH INTERPRETER INTERFACE
The interpreter interface resembles that of the UNIX shell: when
called with standard input connected to a tty device, it prompts for
commands and executes them until an EOF is read; when called with a
file name argument or with a file as standard input, it reads and
executes a
.I script
from that file;
when called with
.B \-c
.I command,
it executes the Python statement(s) given as
.I command.
Here
.I command
may contain multiple statements separated by newlines.
Leading whitespace is significant in Python statements!
In non-interactive mode, the entire input is parsed befored it is
executed.
.PP
If available, the script name and additional arguments thereafter are
passed to the script in the Python variable
.I sys.argv ,
which is a list of strings (you must first
.I import sys
to be able to access it).
If no script name is given,
.I sys.argv
is empty; if
.B \-c
is used,
.I sys.argv[0]
contains the string
.I '-c'.
Note that options interpreted by the Python interpreter itself
are not placed in
.I sys.argv.
.PP
In interactive mode, the primary prompt is `>>>'; the second prompt
(which appears when a command is not complete) is `...'.
The prompts can be changed by assignment to
.I sys.ps1
or
.I sys.ps2.
The interpreter quits when it reads an EOF at a prompt.
When an unhandled exception occurs, a stack trace is printed and
control returns to the primary prompt; in non-interactive mode, the
interpreter exits after printing the stack trace.
The interrupt signal raises the
.I Keyboard\%Interrupt
exception; other UNIX signals are not caught (except that SIGPIPE is
sometimes ignored, in favor of the
.I IOError
exception).  Error messages are written to stderr.
.SH FILES AND DIRECTORIES
These are subject to difference depending on local installation
conventions:
.IP /usr/local/bin/python
Recommended location of the interpreter.
.IP /usr/local/lib/python<version>
Recommended location of the directory containing the standard modules.
.SH ENVIRONMENT VARIABLES
.IP PYTHONHOME
Change the location of the standard Python libraries.  By default, the
libraries are searched in <prefix>/lib/python<version> and
<exec_prefix>/lib/python<version>, where <prefix> and <exec_prefix>
are installation-dependent directories, both defaulting to
/usr/local.  When $PYTHONHOME is set to a single directory, its value
replaces both <prefix> and <exec_prefix>.  To specify different values
for these, set $PYTHONHOME to <prefix>:<exec_prefix>.
.IP PYTHONPATH
Augments the default search path for module files.
The format is the same as the shell's $PATH: one or more directory
pathnames separated by colons.
Non-existant directories are silently ignored.
The default search path is installation dependent, but generally
begins with <prefix>/lib/python<version> (see PYTHONHOME below).
The default search path is always appended to $PYTHONPATH.
If a script argument is given, the directory containing the script is
inserted in the path in front of $PYTHONPATH.
The search path can be manipulated from within a Python program as the
variable
.I sys.path .
.IP PYTHONSTARTUP
If this is the name of a readable file, the Python commands in that
file are executed before the first prompt is displayed in interactive
mode.
The file is executed in the same name space where interactive commands
are executed so that objects defined or imported in it can be used
without qualification in the interactive session.
You can also change the prompts
.I sys.ps1
and
.I sys.ps2
in this file.
.IP PYTHONDEBUG
If this is set to a non-empty string it is equivalent to specifying
the \fB\-d\fP option.
.IP PYTHONINSPECT
If this is set to a non-empty string it is equivalent to specifying
the \fB\-i\fP option.
.IP PYTHONUNBUFFERED
If this is set to a non-empty string it is equivalent to specifying
the \fB\-u\fP option.
.IP PYTHONVERBOSE
If this is set to a non-empty string it is equivalent to specifying
the \fB\-v\fP option.
.SH SEE ALSO
Python Tutorial
.br
Python Library Reference
.br
Python Reference Manual
.SH AUTHOR
.nf
Guido van Rossum
CNRI
1895 Preston White Drive
Reston, VA 20191
USA
.PP
E-mail: guido at cnri.reston.va.us, guido at python.org
.fi
.PP
And a cast of thousands.
.SH INTERNET RESOURCES
Web site: http://www.python.org
.br
FTP site: ftp://ftp.python.org
.br
Newsgroup: comp.lang.python
.SH COPYRIGHT
Copyright 1991-1995 by Stichting Mathematisch Centrum, Amsterdam,
The Netherlands.
.IP " "
All Rights Reserved
.PP
Permission to use, copy, modify, and distribute this software and its
documentation for any purpose and without fee is hereby granted,
provided that the above copyright notice appear in all copies and that
both that copyright notice and this permission notice appear in
supporting documentation, and that the names of Stichting Mathematisch
Centrum or CWI or Corporation for National Research Initiatives or
CNRI not be used in advertising or publicity pertaining to
distribution of the software without specific, written prior
permission.

While CWI is the initial source for this software, a modified version
is made available by the Corporation for National Research Initiatives
(CNRI) at the Internet address ftp://ftp.python.org.

STICHTING MATHEMATISCH CENTRUM AND CNRI DISCLAIM ALL WARRANTIES WITH
REGARD TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS, IN NO EVENT SHALL STICHTING MATHEMATISCH
CENTRUM OR CNRI BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL
DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
PERFORMANCE OF THIS SOFTWARE.

From Sven.Drescher at dlr.de  Wed Aug 18 03:50:55 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Wed, 18 Aug 1999 09:50:55 +0200
Subject: Embedding python
Message-ID: <7pdnso$om6@news.go.dlr.de>

Hi!

I want to embed python's Tk to C++, but I can't compile it. I looked for the
libraries, but I didn't find the right.

I have a Sun with SunOS 5.7 and Solaris CDE Version 1.3.

Now I need the right library, which includes the definition of the regexpr.o
module.

The following modules and libraries I have linkes with it:
- getpath.o
- config.o

- lsocket
- lpython1.5
- lnsl
- lm
- ldl
- lexpect
- lpcre

But once is missing.

Thanks for help and hints!

Sven





From tim_one at email.msn.com  Tue Aug 24 01:45:49 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 24 Aug 1999 01:45:49 -0400
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <7pp3ub$gdj@dfw-ixnews7.ix.netcom.com>
Message-ID: <000501beedf3$eb737200$902d2399@tim>

[Aahz Maruch]
> ...
> Here's another question: I know the *period* of whrandom() is very
> long, but what is the granularity/precision of it?

It combines 3 sequences each of which have no better than 15-bit resolution.
Analyzing the combined effect is tricky.  You need to define your requirements
precisely, then write a realistic simulation independent of your app to test
whether your requirements are met.  Every non-trivial application of
pseudo-random numbers should do that.  Collecting random <wink> info about the
period and resolution is no substitute; e.g., the lattice structure of
random.random may kill you no matter how long its period or fine its
resolution -- or it may be irrelevant.  It depends on every detail of your app.

random.random's prime virtue is that it's fast, not that it's especially good
(see Ivan Frohne for a Python implementation of the Mersenne Twister, which is
a much stronger generator that's slower, but not (like md5) dramatically
slower).

one-hour-of-you-testing-will-save-me-two-days-of-thinking-
    and-is-a-lot-more-trustworthy-too<wink>-ly y'rs  - tim





From ivanlan at callware.com  Tue Aug 31 11:52:58 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 31 Aug 1999 15:52:58 GMT
Subject: Apache:  new release and nice change
Message-ID: <37CBFA5A.1300931F@callware.com>

Hi All--

I just received the email announcement for the release of Apache 1.3.9,
and noticed this little announcement in the blurb for changes:

> - Script errors on Windows are now copied to the server's error log
>   if the scripts fail; this is closer to the Unix behaviour.

That should help debugging a little.  The Apache team recommends that
you upgrade to 1.3.9.  I'll be doing it today sometime.

<remember-python-is-a-scripting-language>-ly y'rs,
Ivan;-)
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From someone at someplace.com  Fri Aug 27 20:19:49 1999
From: someone at someplace.com (Someone)
Date: Sat, 28 Aug 1999 00:19:49 GMT
Subject: Why do we call python a scripting language?
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com>
Message-ID: <37c726c4.2148316@news.netset.com>

On 27 Aug 1999 12:59:18 GMT, fredp at multimania.com.nospam (Fred
Pacquier) wrote:

>for the profane. The one app that 'did it' for me, curiously, is PySol. 
>It's a game, but a game engine, and a powerful one. And it shuffles around 
>a lot of graphics (pun intended), which may surprise people biased by the 
>'scripting language' tag.

This brings me to a new point.  since I first used python, I've been
wondering about distribution of my final products.  In general, i dont
want to distribute a full python distribution and all my source, as
much from not wanting to burden the user with this task as for
protecting my source.  I've played with freeze, but since I never
really had anything to distribute, I've never really used it for
anything except play things.  But from what I remember, there is
nothing preventing a person from putting into a small package
everything needed to distribute an app.

But what I find most interesting is that I havent seen much evidence
that any one uses it.  Is any one using freeze or some derivative or
work alike?  Most things I find are written for users of python, and
are just distributed as source.

>Yet another selling point might be if IDLE or PythonWin or some other seed 
>came to evolve into a full-blown IDE up to par with Microsoft's or 
>Borland's, that might impress the developer population. After all, VC++ is 
>done in MFC/C++ and Delphi in ObjectPascal, right ? But I seem to recall 
>this was all discussed to death some months ago... :-)

I'm very impressed with what I am seeing on the python win 2 beta
site.  its very impressive!  

-guy



From da at ski.org  Mon Aug  2 02:44:22 1999
From: da at ski.org (David Ascher)
Date: Sun, 1 Aug 1999 23:44:22 -0700 (Pacific Daylight Time)
Subject: New version of compile.py
Message-ID: <Pine.WNT.4.04.9907290939080.159-100000@rigoletto.ski.org>

I've upgraded the version of compile.py on my website.  It is now at:

  http://starship.python.net:9673/crew/da/Code/compile

What it is:

compile.py is a tool which reads many Setup files and automates the
building of Python extensions on Windows.  It incorporates a modified
version of Konrad Hinsen's compile.py which worked only on Unix.

What's new:

There have been many changes since the last website release.  Most
significantly, it now automatically builds the extensions using PythonCOM
and Visual Studio 6.0, if both of those are available.  Also, it
automatically creates projects for both Debug and Release configurations.

Cheers,

--david

PS: The entire website has been redesigned.






From paul at digicool.com  Fri Aug 13 08:08:29 1999
From: paul at digicool.com (Paul Everitt)
Date: Fri, 13 Aug 1999 08:08:29 -0400
Subject: Zope for python.org?
References: <7ouecd$9m$1@nnrp1.deja.com> <14259.30266.418140.625710@anthem.cnri.reston.va.us>
Message-ID: <37B40ABD.2C5F8460@digicool.com>

bwarsaw at python.org wrote:
> 
> >>>>> "DCG" == Dinu C Gherman <gherman at my-deja.com> writes:
> 
>     DCG> Would it be desirable to turn python.org into a Python
>     DCG> "portal site" using the increasingly popular Zope product?
> 
> I'm very interested in looking at Zopifying python.org at some point,
> probably after Zope 2.0 is released.  If it helps better serve the
> Python community by allowing a more direct distributed management of
> the site, and coincidentally makes my job as webmaster easier, I'm all
> for it.

Yes, we've had some discussions on it.  Our goal here is to get Zope2
and the portal toolkit stuff running and useful on the new
soon-to-be-announced zope.org site.  I think it shows promise, and will
provide a platform for collecting a lot of ideas about content, but
right now the glitches would drain way too much of Barry's time.

September is a good time to start thinking more seriously about it. 
We'll probably discuss it at DCPIGgie.

--Paul



From AKiss at GEOMETRIA.hu  Tue Aug  3 04:31:39 1999
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Tue, 3 Aug 1999 10:31:39 +0200
Subject: stuck on COM interface to Word
Message-ID: <01BEDD9B.6052EFB0@exchange1.geometria.hu>

Hi Jim,
Here is an example:
	import time
	print time.ctime(int(doc.BuiltInDocumentProperties("Last Print Date")))
The online help contains other useful keywords(find the details of the
DocumentProperty): "Title","Subject","Author" etc.
Regards,
Arpad


Jim Kerr <jbkerr at sr.hp.com> wrote in message
news:<37A60515.370B5085 at sr.hp.com>...
>   I'm trying to write a python script to extract some properties from a set
> of Word documents. The attributes I'm interested in are things like author's
> name, last print date, keywords, etc.
> 
>   This is what I have so far:
> 
> from win32com.client import constants,Dispatch
> import string,os
> 
> app = Dispatch('Word.Application.8')
> app.Documents.Open(r"C:\My Documents\myMemo.doc")
> doc = app.ActiveDocument
> print "Last printed date: ",  # ?? what goes here ??
> 
>   I tried using combrowse.py to figure out what the properties of the
doc are,
> but I couldn't find anything that looked relevant.
> 
>   Is there any web-based info on COM interfaces to various programs? Or
do you
> have to buy Visual Basic books and translate for yourself?
> 
>   Thanks for the help!
> 
> 						Jim
> -- 
> Jim Kerr
> Hewlett-Packard
> Santa Rosa Systems Division
> 1400 Fountaingrove Pkwy, MS 3USZ
> Santa Rosa, CA 95403
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From dfan at harmonixmusic.com  Tue Aug 17 15:25:57 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 17 Aug 1999 15:25:57 -0400
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <14265.35455.590852.517197@bitdiddle.cnri.reston.va.us>
Message-ID: <wkyafagqey.fsf@turangalila.harmonixmusic.com>

Jeremy Hylton <jeremy at cnri.reston.va.us> writes:

| I've rather fond of the following approach:
| 
| import random
| 
| def randomize(a, b):
|     return random.choice([-1, 0, 1])
| 
| l = range(2, 20)
| l.sort(randomize)

I'm not sure how robust this is.  I've seen at least one implementation
of qsort crash when given an inconsistent comparison function (e.g.,
one that doesn't obey transitivity).  And how is sort() supposed to
know when it's done?

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From amitp at Xenon.Stanford.EDU  Mon Aug 23 15:03:03 1999
From: amitp at Xenon.Stanford.EDU (Amit Patel)
Date: 23 Aug 1999 19:03:03 GMT
Subject: RPM for Python 1.5.2 ?
References: <slrn7qrn3u.7fj.dyoung@baltazar.baltazar.org> <37ADE4C1.194C31D6@lemburg.com> <19990809113731.A6757@baltazar.baltazar.org> <37AFD5A9.4C45332F@lemburg.com>
Message-ID: <7ps5t7$2u6$1@nntp.Stanford.EDU>

| > On Sun, Aug 08, 1999 at 10:12:49PM +0200, M.-A. Lemburg wrote:
| > > David Young wrote:
| > > >
| > > > Has anyone built an RPM for Python 1.5.2 ? I think it's sort of curious
| > > > that RedHat hasn't already. Anyway, I'm building my own package right now,
| > > > but I'm afraid I might be wasting my time....
| > >
| > > Try:
| > >
| > >       http://www.andrich.net/python/download.html


I had good luck with these RPMs on RH 6 except  when I tried
installing the Tkinter stuff:

  # rpm -Uvh python-_tkinter-1.5.2-2.i386.rpm 
  error: failed dependencies:
          blt >= 2.4c is needed by python-_tkinter-1.5.2-2
          libBLT.so.2 is needed by python-_tkinter-1.5.2-2

I don't know what blt is!  I did erase RH's tkinter package, but it
didn't help.  Do I need to install Tcl/Tk extensions to make it work
with Python?  :(  

For normal use I don't use Tkinter so it's not affecting me.  All I
wanted to do with Tkinter was to try out IDLE.. :) I'm quite happy
with Oliver's RPMs.. it's much more convenient than asking everyone at
work to compile Python 1.5.2 on their system. :)

     - Amit



From bogw at geo.vu.nl  Wed Aug 18 12:50:19 1999
From: bogw at geo.vu.nl (Patrick Bogaart)
Date: Wed, 18 Aug 1999 18:50:19 +0200
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net>
Message-ID: <37BAE44B.E66250BD@geo.vu.nl>

Ben Caradoc-Davies wrote:
> 
> [strange programming]
> The most confidence uninspiring piece of code I ever had the displeasure to
> maintain contained something like this (allegedly C++, but mainly it's
> intersection with C):
> 
>         i = 0;
>         while( i < max_index ) {
>                 /* some copying or update (not modifying i) goes here */
>                 i = i + 1;
>         }
> 
> This was written by a *very* expensive consultant working for a large
> multinational accountancy firm.
> 
> Can anybody suggest an alternative construct which more accurately expresses
> the programmer's intentions? There may be more than one way to do it, but,
> puhleeze!
> 

Yes I can:

	i = 0;
	label_1:;
	/* some use of i */
	i = i + 1;
	if (i==max_index) goto label_2;
	goto label_1;
	label_2:;

<wink>

--
Patrick W. Bogaart
Department of Geomorphology and Quaternary Geology
Faculty  of Earth Sciences,  Vrije Universiteit Amsterdam
bogw at geo.vu.nl            http://www.geo.vu.nl/users/bogw



From arcege at shore.net  Sun Aug 15 21:15:40 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 16 Aug 1999 01:15:40 GMT
Subject: C's isprint() concept?
References: <slrn7rea88.9uf.jblaine@shell2.shore.net> <Pine.GSO.4.10.9908151713380.3761-100000@crusoe.crusoe.net>
Message-ID: <0DJt3.195$mV.76681@news.shore.net>

Jeff Pinyan <jeffp at crusoe.net> wrote:
:> If I want to replace all non-printable characters in a string with a single
:> space, what would be the best way?  Do I need to loop over the entire string
:> character by character checking the ord() value of each one?  Anyone have
:> a sane way to do this with regular expressions?

: In Perl, I could do it this way:
:   $string =~ tr/\x00-\x1f\x80-\xff//d;

: What that means is this:
:   remove the following characters from $string:
:     characters whose ASCII value is from \x00 (0) to \x1f (31)
:     characters whose ASCII value is from \x80 (128) to \xff (255)

: This can be done, almost as painlessly, in Python.

Considering this was asked to a Python newsgroup, how about showing
how to do it in Python.

  import string
  # works for ASCII
  control_chars = string._idmap[:ord(' ')]  # 0 to 31
  high_chars = string._idmap[ord('~')+:]    # 127 to 255
  to_remove = control_chars + high_chars
  map = string.maketrans(to_remove, ' ' * len(to_remove))

  midstr = string.translate(instr, map)
  outstr = string.join(string.split(midstr))

or:
  import re
  outstr = re.sub(r'[^ -~]+', ' ', instr)

(Indented for "easy cut&paste" ;)

  -Arcege




From bradh at mediaone.net  Sat Aug 14 22:37:08 1999
From: bradh at mediaone.net (Brad Howes)
Date: 14 Aug 1999 22:37:08 -0400
Subject: Wacky Programming Tales (Was Re: Why use Perl when we've got Python?!)
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net>

Sorry to intrude on this love fest...

sholden at pgrad.cs.usyd.edu.au (Sam Holden) writes:
<snip>

> It's like someone who starts a C program with :
> 
> #define BEGIN {
> #define END }
> 

Hey! I actually worked on a project where someone did this! Also in their
funky include file was

  #define AND &&
  #define OR  ||

It was the most bizarre thing I ever saw -- not so much the include file,
but the resulting code. As if a Pascal moths had come in, eaten at some
code and left little Wirth fragments sprinkled around. Has anyone else
encountered such strange programming behavior?

...now back to my language is bigger/longer/wider/stiffer than yours

-- 
Brad Howes                    bradh at mediaone.net



From news at dorb.com  Sat Aug 14 19:29:39 1999
From: news at dorb.com (Darrell)
Date: Sat, 14 Aug 1999 19:29:39 -0400
Subject: deadlock when two servers call each other - will threads help?
References: <199908142158.QAA24395@dolphin.mojam.com>
Message-ID: <MZmt3.1023$8P.26672@typhoon.nyroc.rr.com>

There is a lot of stuff written about avoiding dead lock.
Adding threads can help, but also add more stuff that can dead lock as well
as complexity. Anything I would be able to suggest will most likely impact
your architecture. But here goes.

My favorite solution is message queues that prevent the sender from
blocking. The receiver is blocked reading the message queue. The hard part
is that the receiver is running more like a dispatcher. Messages must be
matched up with callbacks that handle that kind of message.

The handler can register for messages based on all kinds of criteria, and
time outs.

This is about how Microsoft's message queues work and I suspect how the
Medusa web server works.

If you want to know more I can dig it up.
Also know as the "reactor pattern", I think ??
--
--Darrell

Skip Montanaro <skip at mojam.com> wrote in message
news:199908142158.QAA24395 at dolphin.mojam.com...
>
> I'm using XML-RPC as the transport mechanism within a group of database
> servers to propagate change information and thus maintain database
> consistency.  My servers are logically connected in a tree structure.
Each
> server has one parent (except the topmost server) and zero or more
> children.  When a change request comes in to a server, if it doesn't come
> from the server's parent, it forwards it there.  This allows requests to
> come in anywhere in the system, then get percolated through all the
servers
> by forcing it to travel to the top of the tree, then trickle down.
>
> This is fine in theory, but in practice it yields deadlocks.  A child can
be
> making an XML-RPC call to its parent to forward one change request at the
> same time the parent is trying to make another call to that child.  Both
> servers will block in read() waiting for a response that never arrives.
> System throughput suffers...
>
> I don't use threads at the moment, but I'm curious whether or not threads
> would help here.  Would the blocked read() call allow another thread to
run?
> If so, I could spawn threads to make RPC calls between servers.
>
> Thanks,
>
> Skip Montanaro | http://www.mojam.com/
> skip at mojam.com  | http://www.musi-cal.com/~skip/
> 847-971-7098
>
>
>





From tim_one at email.msn.com  Sun Aug 22 03:11:48 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 22 Aug 1999 07:11:48 GMT
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <005201beea1a$5ab05f30$f29b12c2@secret.pythonware.com>
References: <005201beea1a$5ab05f30$f29b12c2@secret.pythonware.com>
Message-ID: <000101beec6d$99d1b380$542d2399@tim>

[Tim, on Python's sort function and its robustness against sick
 inputs]

[Fredrik Lundh]
> how about JPython

Don't know.

> and Viper

*Nobody* knows <wink>.

> and Lumberjack

Ditto.

> and SPython and FooPython and whatever...

Ditto ditto ...

> as far as I can tell, the "spec" says
>
>     sort: sort the items of s in place
>
> and that's about it...

In case of conflict, CPython usually wins.  Is it your contention that
because the spec doesn't say sort must not dump core, therefore it's OK if
it does?  If so, the reference manual is going to need one hell of a lot
more words all over the place <wink>.

in-std-speak-this-kind-of-stuff-is-swept-under-the-"quality-
    of-implementation"-rug-ly y'rs  - tim





From goertzen at norscanNOSPAM.com  Mon Aug 30 10:35:44 1999
From: goertzen at norscanNOSPAM.com (Daniel Goertzen)
Date: Mon, 30 Aug 1999 14:35:44 GMT
Subject: running python on a floppy?
References: <z6ex3.1541$1i4.128395@typhoon.mbnet.mb.ca> <37CA50EC.4FC03560@home.com>
Message-ID: <4Fwy3.2921$1i4.189540@typhoon.mbnet.mb.ca>

Doug Hellmann <doughellmann at home.com> wrote in message
news:37CA50EC.4FC03560 at home.com...
> Why are people still trying to fit distributions like this on a
> *floppy*?  Why not use a Zip disk?  Granted, they are more expensive,
> but there is *so* much more room.

I did try the zipslack distribution, but the continuous stream of kernel
warnings didn't encourage me.  Also, zip drives are not the most reliable
devices on the face of the earth (not that floppies are much better), and
are relatively expensive.

After looking at the problem further, I've come to conclusion that 1.44 mb
is just too tight.  An 8 meg flash disk with a compressed filesystem should
give me about 30 megs of space to play with; more than enough to do lots of
interesting things with python.

Regards,
Dan.






From aahz at netcom.com  Sat Aug 21 15:36:51 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 21 Aug 1999 19:36:51 GMT
Subject: can some one help a newbie?
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <7pkke6$lut@dfw-ixnews6.ix.netcom.com> <935227166snz@vision25.demon.co.uk>
Message-ID: <7pmv4j$gtt@dfw-ixnews14.ix.netcom.com>

In article <935227166snz at vision25.demon.co.uk>,
Phil Hunt <philh at vision25.demon.co.uk> wrote:
>In article <7pkke6$lut at dfw-ixnews6.ix.netcom.com>
>           aahz at netcom.com "Aahz Maruch" writes:
>>
>> Overall, if your background is as a sysadmin using shell scripts, grep,
>> awk, sed, and so on, you'll find Perl a bit easier.
>
>I've used all these, and find Python easier than Perl. Perl would
>probably be easier than Python for 1-liners, if I could remember
>the syntax.

Note that I'm specifically assuming a sysadmin with little actual
programming experience.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From sholden at pgrad.cs.usyd.edu.au  Sat Aug 14 22:10:03 1999
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 15 Aug 1999 02:10:03 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au>  <37B54DA3.4210DC10@nospam.com>
Message-ID: <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au>

On Sat, 14 Aug 1999 12:06:11 +0100, Alex Maranda <amaranda at nospam.com> wrote:
>Sam Holden wrote: 
>> >
>> >>You can write Perl programs that resemble sed, or awk, or C, or Lisp, or
>> >>Python. This is Officially Okay in Perl culture.
>> >
>> >You cannot write Perl that resemble Python.  You are required to
>> >use curly braces as block delimiters.
>> 
>> In fact you are not. You are wrong yet again. Actually thinking about how
>> perl works for a minute would show you that perl is in fact flexible.
>> it is not a lie when people say perl doesn't force you to write a
>> particular way.

<snip impythonate.pm by Damian Conway>

>> And here is his sample code that is now valid perl (although anyone who uses
>> it for real code should be killed) :
>Why? You don't happen to hate Pythoneers or Python itself, do you? I've
>actually appreciated the argument from your second post, when you taxed
>the other guy for needlessly questioning your Python knowledge out of
>context.

I know a fair few pythoners in fact, I even like some of them. What I meant
was that anyone who would do such a thing with perl deserves to be killed.
Since no one else will be able to maintain the program. It's like someone
who starts a C program with :

#define BEGIN {
#define END }

Doing things like this means that a C programmer can't understand the code,
and a Pascal programmer can't either. With impythonate.pm, a perl
programmer can't understand the code, and a python programmer can't either.

Also don't tell anyone but I actually like python a lot. I don't use it much
since I have more experience with perl, and more importantly I know the
modules available for perl better, and can quickly search for one to
do whatever I need next.

>I am completely ignorant with Perl and I can barely read what's above
>(by interpolating my knowledge of shell, C, and Tcl).

That's because it uses a lot of the not so well known perl features. 

>
>> 
>> #Now newlines replace colons and indentation replaces brackets:
>> 
>> use impythonate; # STILL NEED THAT ONE SEMICOLON, DAMMIT!
>> 
>> for $i (1..10)  # COMMENTS ARE OKAY
>>   print "$i: "
>>   my $isq =  \
>>   $i **2 # LINE CONTINUATIONS WORK AS IN PYTHON
>>   print " $isq\n"
>> 
>> print "done\n"
>I know Python fairly well (but I'm no expert) and I can easily read this
>piece of Perl.

But it would be no harder to read with ; on the end of each statement and
a pair of {}s.

Working out that strings interpolate should be a bigger leap than working
out that blocks have {}s around them.

> Getting back to the killing part, which kind of disturbs
>me when reading comp.lang.python:
>- was that necessary to make a point?
>- if yes, what was the point?
>- if the point (just guessing) was that Perl code should never look like
>Python, isn't that against of 'there should be more ways of doing it?'

I answered this above a little.

It was necessary since I don't want someone actually using such an untested
module in their code. Chances are it breaks on many python constructs. The
obvious example is :

if (1==1) : print "fred"

It won't handle that.

Everyone seems to misunderstand the whole concept of TMTOWTDI. It doesn't
mean all ways are equal. That is a way of writing perl which works, it is
however not very maintanable. The only way to determine what syntax is
valid is to know perl's syntax, and to be able to read and understand the
impythonate.pm code - since there is no documentation and it is _not_ the
same as python.

Writing it in 'normal' perl reduces the requirements to understanding, and
speeds it up to boot.

>I fail to see the interesting part in this whole thread, which disturbs
>the habitat of comp.lang.python. I think John Stevens was mistaken to
>bring the discussion over to this newsgroup in the first place. We don't
>have strong feelings about Perl, and don't have plans for world
>domination or anything.
>
>A couple of days ago I read a bit in Bugzilla's README which amused me:
>"A computer which doesn't have Perl installed is a very sad computer
>indeed". Your post is frightening (unless it was a Perl-style joke).

How was it frightening?

If you want to program in python, then use python. It's a nice language,
it is portable, it is simple to install, it is easy to learn. 

So what is the point of munging another language to look like it? If you could
munge perl so that you could name the capturing elements of a regex then that
would be a useful, python like addition...

-- 
Sam

There's no such thing as a simple cache bug.
	--Rob Pike



From mstenber at cc.Helsinki.FI  Mon Aug 30 06:41:13 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 30 Aug 1999 13:41:13 +0300
Subject: Upper limit?!
References: <37CA1E0F.FA2A05A3@megsinet.net> <014f01bef2bd$3fd09e90$f29b12c2@secret.pythonware.com> <37CA3EBD.4077AF18@megsinet.net>
Message-ID: <al8aer9czzq.fsf@sirppi.helsinki.fi>

David Oppenheimer <davidopp at megsinet.net> writes:
> Wow, that's quite a difference! on an Alpha you can have a 19 digit number
> as opposed to a 10 digit number on my Pentium.  So the limit isn't actually
> a factor of the programming language but a virtue (or limitation) of my
> processor.  I wonder if Intel knows about this?!   LOL  Makes me want to

Yes, for machine-spesific ints.. However, some languages can kludge around
it, like Python's bigints:

fingon at dreamland ?uname -m
i686
fingon at dreamland ?python
Python 1.5.2 (#0, Jul 16 1999, 19:02:32)  [GCC egcs-2.91.66 Debian
GNU/Linux (egcs-1.1.2 release)] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> len(str(2L**666))
202
>>>

202 digits, baby ;-)

> sell this hunk of crap Packard Hell P-75, get rid of Win98 and buy an Alpha
> running Linux.   That PSN number thing scares the hell outta me any ways,
> I'm either going AMD or Alpha next go around.

Heh.

> David O.

-Markus Stenberg

-- 
 Running Windows on a Pentium is like having a brand new Porsche but
 only be able to drive backwards with the handbrake on.
	(Unknown source)



From mcfletch at vrtelecom.com  Mon Aug  9 18:13:15 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Mon, 9 Aug 1999 22:13:15 GMT
Subject: Packing and unpacking IP addresses into 32-bit integers
In-Reply-To: <Pine.A32.3.90.990809163915.17213C-100000@elvis.med.Virginia.EDU>
References: <Pine.A32.3.90.990809163915.17213C-100000@elvis.med.Virginia.EDU>
Message-ID: <NDBBIAPHEKGOBIHCBLIICEDKCAAA.mcfletch@vrtelecom.com>

Wouldn't it be something like:
	'>bbbb' or '!bbbb'

instead (for portability to non-big-endian machines (I'm assuming the packed
representation is supposed to have network byte order))?

Just a thought,
Mike

-----Original Message-----
From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
Behalf Of Steven D. Majewski
Sent: August 9, 1999 4:50 PM
To: python-list at cwi.nl
Subject: Re: Packing and unpacking IP addresses into 32-bit integers

...
and use either struct.pack or array to coerce it into a binary string:

  a = apply( struct.pack, [ 'bbbb' ] + adr )




From mpersico at erols.com  Sun Aug 15 15:15:17 1999
From: mpersico at erols.com (Matthew O. Persico)
Date: Sun, 15 Aug 1999 15:15:17 -0400
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net>
Message-ID: <37B711C5.32C1CA6C@erols.com>


Ben Caradoc-Davies wrote:
> 
> [strange programming]
> The most confidence uninspiring piece of code I ever had the displeasure to
> maintain contained something like this (allegedly C++, but mainly it's
> intersection with C):
> 
>         i = 0;
>         while( i < max_index ) {
>                 /* some copying or update (not modifying i) goes here */
>                 i = i + 1;
>         }
> 
> This was written by a *very* expensive consultant working for a large
> multinational accountancy firm.
> 
> Can anybody suggest an alternative construct which more accurately expresses
> the programmer's intentions? There may be more than one way to do it, but,
> puhleeze!
> 
> Idiomatic is not quite the word.

Try idiotic!

> --
> Ben Caradoc-Davies <bmcd at es.co.nz>

-- 
Matthew O. Persico
    
You'll have to pry my Emacs from my cold dead oversized
   control-pressing left pinky finger. -- Randal L. Schwartz



From andrew at malcolmson.com  Wed Aug 11 19:56:20 1999
From: andrew at malcolmson.com (Andrew Malcolmson)
Date: Wed, 11 Aug 1999 19:56:20 -0400
Subject: internet programming with python
References: <37AF678E.CC8E165B@netscape.com> <37B1AB33.1C81BDEA@am.appstate.edu>
Message-ID: <37B20DA3.3E3ABD98@malcolmson.com>

I also have it "Internet Programming with Python" and found it very good, fun
to read, and probably a better book for a Python beginner than "Learning
Python".  I hope it gets updated.

However, are you saying that there is something in it about using Python with
Netscape Enterprise Server?  I didn't see anything there.

TMGB wrote:

> I just picked it up at Barnes and Noble where it retailed for $34.95 US
> and is discounted to $17.48 US.  I feel sure that you can probably order
> it from http://www.barnes&noble.com (not sure about URL).  The ISBN is
> 1-55851-484-8 published by M&T Books a division of MIS:Press,
> INC(http://www.mandt.com).  I'm not sure but I think some references are
> to libraries that have now been updated or replaced but in general it
> has been a very helpful introduction.  Especially interesting (to me),
> if you have Netscape Enterprise Server, is embedding Python into
> Netscape Server NSAPI.
>
> Thomas
>
> Vijay Mallela wrote:
> >
> > Hi,
> > I am looking for the book "Internet programming with python". I need it
> > at the earliest..
> > I checked out amazon and other stores and they dont have it.. does any
> > one know where I can get the book
> > Thanks
> > vijay
>
> --
> ----------------------------------------------------------------------
> Thomas McMillan Grant Bennett           Appalachian State University
> Computer Consultant II                  University Library
> bennettt at am.appstate.edu
> http://www.library.appstate.edu/admin/
> Voice:  828 262 6587                    FAX:    828 262 3001
>
> Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
> operating system that was originally coded for a 4-bit microprocessor.
>  - Chris Dunphy     Boot Magazine




From cjw at connection.com  Mon Aug  2 07:28:20 1999
From: cjw at connection.com (Colin J. Williams)
Date: Mon, 02 Aug 1999 07:28:20 -0400
Subject: Call for snippets
References: <199908020821.KAA07953@axil.hvision.nl>
Message-ID: <37A580D4.202AF94F@connection.com>

Hans,

Thanks for this posting, the site is clean and well organized.

It is a useful learning tool.  I might, one day, make an offering.

One little suggestion, you might consistently date the entries.

Colin W,

Hans Nowak wrote:

> Hello,
>
> Some of you may have noticed that I have a snippets site since a few
> months. (In this definition, snippets are pieces of working Python
> code, usually demonstrating something or solving a problem.) Until
> now, I have extracted them from newsgroup messages. This isn't an
> ideal situation, because there are lots of messages which I all have
> to read, and I don't really have the time for that. (Yeah I know...
> sacrilege... :^)
>
> This isn't really a problem though. What does bother me is that my
> collection isn't growing very rapidly. I have 119 snippets online
> now, with 100+ still residing on my harddisk, untested. It's a start,
> but there really isn't much progress, mainly because of the reasons
> mentioned above. I have a feeling that there is a lot of good and
> useful code there but that I have to wait until it accidentally shows
> up in the newsgroup, triggered by somebody's question.
>
> So, my request is obvious. I would ask everyone to browse through
> their code and see if they have any interesting, useful or just plain
> cool snippets. I'm not really picky; a nifty function would do. I am
> not really looking for complete modules since I have a snippet
> repository, not a module repository (nor anything like CPAN, the
> pyPit, etc), but I won't refuse them. In fact, anything that solves
> any problem could be a valuable snippets. I'd almost say, 'when in
> doubt, post your code'. <g>  I'd go even farther: if you have Python
> code somewhere and think there might be interesting functions in it,
> I would be pleased if you pointed me to it; I will scan and extract
> them myself.
>
> My goal is to have a site where both newbies and more advanced coders
> could go to see how to do something in Python, or how to do something
> *differently* if they already have some solution, because more
> snippets can tackle the same problem. But in order to do that I need
> some more code! Or at least a larger base. <wish> Now if everyone
> reading this ng would post at least one snippet, there would be an
> explosive growth... </wish>
>
> ?ou can check out the page at http://www.hvision.nl/~ivnowa/snippets/
> . It's (of course) still under construction. Snippets and comments
> can be sent to (feeble spam protection) ivnowa @ hvision . nl or
> hnowak @ cuci . nl. (The address mentioned on the side is actually
> obsolete. I will remove it.)
>
> Thanks in advance,
>
> Hans Nowak (ivnowa at hvision.nl)
> Homepage: http://fly.to/zephyrfalcon




From mstenber at cc.Helsinki.FI  Sun Aug  1 07:24:38 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 01 Aug 1999 14:24:38 +0300
Subject: NT select.select?
References: <613145F79272D211914B0020AFF6401914DC19@gandalf.digicool.com>
Message-ID: <al8d7x7216x.fsf@sirppi.helsinki.fi>

Brian Lloyd <Brian at digicool.com> writes:
> > Is there some low limit on maximum number of sockets you can 
> > have in the
> > Python-NT's select call? A program that happens to work 
> > perfectly on Linux
> > seems to die on NT around 64(?) sockets to the 'too many file 
> > descriptors
> > in call' error.
> > 
> > Any portable ways to bypass it?
> > 
> > -Markus
> It turns out that NT has a default 64 fd limit on arguments to
> select(). The good news is that you can actually bump the limit up 
> to whatever number you want by specifying a define when compiling 
> python15.dll.

I was hoping for version that works with 'Python out of the box' (I kludged
something in pure python with threads that bypasses this limit, but it
isn't pretty).

> The default 64 fd limit is too low for anyone trying to run
> an async server that handles even a modest load, so I've 
> submitted a bug report to python.org asking that the define
> above find its way into the next python release...

Agreed. Unless it causes some Huge Load(tm) somewhere, it should be
definitely raised. (say, 1024 or 4096 sounds reasonable to me)

> Brian Lloyd        brian at digicool.com
> Software Engineer  540.371.6909              
> Digital Creations  http://www.digicool.com 

-Markus

-- 
"...very few phenomena can pull someone out of Deep Hack Mode, with two
noted exceptions: being struck by lightning, or worse, your *computer*
being struck by lightning."
	- Matt Welsh



From mwh21 at cam.ac.uk  Sun Aug 15 18:10:05 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 15 Aug 1999 23:10:05 +0100
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <m3vham44g5.fsf@atrus.jesus.cam.ac.uk> <37b6e9c3.345781227@news.triode.net.au>
Message-ID: <m3pv0o3dc2.fsf@atrus.jesus.cam.ac.uk>

skaller at maxtal.com.au (John (Max) Skaller) writes:
> On 11 Aug 1999 18:23:06 +0100, Michael Hudson <mwh21 at cam.ac.uk> wrote:
> >I'm interested (do you have any code yet?
> 
> 	Not for the compiler: I'm implementing
> the interpreter first. [I have done some work on the
> type inference algorithm, to make sure it is
> feasible]

Let me rephrase that a little: do you have any code that I can look
at/play with/beat on/improve (maybe)? Actually working/complete code
is not expected or desired, I'm just nosey.

> >I presume fiddling with sys.exc_traceback.f_back.f_locals is out?
> 
> 	At present sys.exc_info() will return a tuple in which
> the third element is a traceback object. At present, that
> traceback object doesn't match the Python one --
> it just has a filename and line number in it, to allow the
> traceback to be printed.

I feel I ought to point out that I'm not particularly desperate to
retain the ability to perform this particular stunt, I was just trying
to get a feel for the limitations you were talking about.

> >But can foo then have a method set_bar? I guess that would be easier
> >to cope with for the analyser.
> 
> 	Yes. It is OK to mutate objects, as opposed
> to rebinding variables: mutation has to be supported.
> You may think these things are equivalent. Indeed,
> they are in interpreted python. 

I'd dispute that, but never mind.

> However, the restriction leads to a performance gain; without some
> restrictions, compiled python will not execute faster than the
> bytecode interpreter.

Quite.

> >> 	5) What restrictions can you live with?
> >
> >One thiung I occasionally find very handy is the ability to assign
> >methods to classes (or more usually within classes).
> 
> 	OK. I would like to do what Guido has done in the
> optimisation of functions [fast lookup for statically known variables,
> plus a dictionary to support others]: provide an optimised structure
> where possible, and then provide a dynamic one
> for the other cases.  That is, the ideal would be to provide
> a 'fallback' to the interpreter.

vtables might be nice (i.e. attribute accesses don't always go through
a dictionary lookup). Bigger win for compiled code, I suspect.

You could only vtable-ize the methods/data members that static
analysis found were never assigned to in code. This suggests another
restriction for exec.

[snipped dictionary extensions that I still don't see the need for but
then again if I don't like them I can just ignore them]
 
> >> 	5) optionally typed function parameters
> >
> >Out of curiosity, what syntax do use for this? 
> 
> 	My first idea was the obvious:
> 
> 	def f(a:int): ..
> 
> or perhaps
> 
> 	def f(a: IntType): ...
> 
> and the interpreter semantics would be 
> 
> 	if type(a) is not InType: raise TypeError, "Int expected for arg1 of f"

Fair enough; it might be nice (& probably very easy) to hack a CPython
that accepted but ignored type declarations of this sort.
 
> However, in the compiled version, the semantic might be to generate
> a compiler error: that is, the following might not work:
> 
> 	try: f(1.0)
> 	except TypeError: pass # just skip the call

I'd say if you coded like that you deserved to lose anyway. Might
become a bigger problem in more complex cases, of course.

> >I presume if a
> >parameter is typed as `Foo' then a value of type `Bar' where Bar
> >derives from Foo is allowable.
> 
> 	I guess that would make sense (the extension isn't
> implemented yet -- not a lot of point appealing for comments
> _after_ implementing the extension, only to find someone
> has a better idea)

I think it would be a hanging offense for anything that called itself
OO to not allow a instance of a derived class to be substituted for
that of a base class.

> >> 	6) what do you want?
> >
> >Can you derive from basic types? That might make some stuff harder.
> 
> 	Not at present. Can you give an example where
> it is useful? [It can be made to work I think: is it useful?]

It's more neat than useful, as far as I can see. But I'll have to
learn more ml, Haskell, smalltalk, Eiffel and lisp/clos before I can
comment properly on type issues, I think (I am actually moderately
serious about learning all the above languages - doing Haskell, Eiffel
and ANSI common lisp at the moment).

> >> 	It is difficult to optimise individual
> >> modules. It is a different story to optimise
> >> a whole program, where _every_ call
> >> to a function can be traced. Of course,
> >> this may involve restricting what 'exec' can do,
> >
> >Like banning it entirely, I suspect.
> 
> 	No. I cannot do that because the program
> that I want to compile, interscript, uses exec,
> and depends on it to execute client script.
> This is the major feature of interscript, so exec
> has to work (at least with restrictions).

I see; I guess what you're saying is that exec-ed code can't muck with
it's calling environment too much. I'd have thought this would be a
feature for interscript, otherwise malicious documents could seriously
injure the environment.

> 	That means the compiled code has to
> contain a full run time system (unless it can be
> detected that it isn't required).

Lisp manages to have a very clean interaction between compiled and
interpreted code: you type (compile 'f) and hopefully the only
noticeable difference is that `f' goes quicker. Mind you lisp has had
man-millennia of time from very very clever people poured into it, so
that might be aiming a bit high ;-).

[snip]

> >Mind you, the whole thing sounds sooo hard, that
> >anythong that makes it easier is good.
> 
> 	The things that turn out to be hard are
> often surprising. 

That's always the way. I think a large part of programming experience
(of which I don't have a vast amount) is spotting the problems and
being able to come up with top-level designs that are compatible with
further growth. Hard problems.

> For example, right now I having
> major problems with something that sounds trivial:
> getting the interpreter to set the line number
> so that a traceback can print out as it does in Python.
> 
> 	That sounds easy, but the LR grammar I'm
> using is not well adapted to it -- NEWLINE tokens
> come after statements, instead of before them,
> where the line number is needed.

Not something I've done, but it's always struck me that
compilers/parsers that only have to deal with good code must be far
easier to right...

Regards,
Michael



From donn at u.washington.edu  Thu Aug 12 18:08:56 1999
From: donn at u.washington.edu (Donn Cave)
Date: 12 Aug 1999 22:08:56 GMT
Subject: Help! $PYTHONPATH trouble
References: <37B33980.A5318181@n-space.org>
Message-ID: <7ovglo$cj0$1@nntp6.u.washington.edu>

Erik Stewart <possum at n-space.org> writes:
| The variable $PYTHONPATH doesn't seem to have any effect on my system (RH
| Linux 6.0, Python 1.5.1).  Here's what I get after defining $PYTHONPATH:
|
| $ PYTHONPATH="home/possum/test/kwp"
| $ echo $PYTHONPATH
| home/possum/test/kwp
| $ python

You need to export the variable into the environment, or it's only a
shell variable that isn't inherited by python or other programs.

  $ PYTHONPATH="home/possum/test/kwp"
> $ export PYTHONPATH
  $ printenv PYTHONPATH
  home/possum/test/kwp
  $ python

Just to make it more confusing, if PYTHONPATH were already an environment
variable you wouldn't have had to export it, and there are several ways
to express that export, too:  export NAME=val,  NAME=val export NAME, etc.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From Doug at news.vex.net  Tue Aug 24 05:05:56 1999
From: Doug at news.vex.net (Lewis)
Date: 24 Aug 1999 09:05:56 GMT
Subject: win32 vim +python
Message-ID: <7ptn9k$mlk$1@hub.org>

Is there a Win32 binary of Vim 5.4(x) compiled with +python anywhere available?
I've looked around, can't seem to find one. If there isn't one... does 
anyone feel like making one for us c-ignorant masses?

Thanks.





From irclark at latveria.castledoom.org  Mon Aug 23 17:00:41 1999
From: irclark at latveria.castledoom.org (Isaac)
Date: 23 Aug 1999 21:00:41 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <slrn7s1cug.1vk.irclark@latveria.castledoom.org> <935437263snz@vision25.demon.co.uk>
Message-ID: <slrn7s3dpt.5a.irclark@latveria.castledoom.org>

On Mon, 23 Aug 99 19:41:03 GMT, Phil Hunt <philh at vision25.demon.co.uk> wrote:
>In article <slrn7s1cug.1vk.irclark at latveria.castledoom.org>
>           irclark at latveria.castledoom.org "Isaac" writes:
>> Very interesting indeed!  I was working on a similar program whose
>> functionality would fit much of the description you give here.  My
>> program reads in a description of a gui that's similar in concept
>> to your par files, and outputs code that implements the gui, in
>> either Motif, GTK, or the EZWGL
>
>What's that?
>
Just some toolkit I found on the net.  I found it useful to have several
different toolkits around in order to make sure the language for the 
backend was general enough to get the job done.  I was hoping to allow
specifying backends by parsing a target file describing the target
tool kit.
>
>My next step, once I'm happy with the parser and the HTML backend,
>will be to write a Python/Tkinter backend. Then, I'll have written two
>backends so I'll have a better idea of how to architect a backend 
>framework -- I want to make backends easy to write, to encourage 
>other people to contribute them.
>
Looks like the right approach.  I think it's important to pick disparate
targets for the first few backends.  GTK with C or C++ might be a good 
choice to add to the ones your already trying.  Not too hard to learn,
and fairly popular.
>
>What language is your tool written in? AFAIK, PCCTS is a C parser
>generator -- is this correct? How does it compare to yacc?
>
PCCTS is a superior tool to yacc/bison.  I don't think I'll ever use yacc
again if I get to pick the tool.  PCCTS generates C or C++.  There is a 
second generation tool, ANTLR, that uses improved technology and generates
java and C++ parsers.

Isaac



From meowing at banet.net  Tue Aug 24 14:49:46 1999
From: meowing at banet.net (meow)
Date: 24 Aug 1999 14:49:46 -0400
Subject: Python equivilent of pack()?
References: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com>
Message-ID: <87r9ktm2t1.fsf@slip-32-100-244-100.ma.us.ibm.net>

Bit Nik <bitwise at io.com> wrote:

> My question is this: is there a functional equivilent to the Perl pack()
> and unpack() functions in Python?

See the standard struct module.  It's very much like the Perl
equivalent, except that the documentation makes sense.



From stuart.hungerford at cmis.csiro.au  Sun Aug 22 20:52:15 1999
From: stuart.hungerford at cmis.csiro.au (Stuart Hungerford)
Date: Mon, 23 Aug 1999 10:52:15 +1000
Subject: Anyone used Python for testing and instrumenting VB code?
Message-ID: <7pq5kc$su0@gutter.its.csiro.au>

Hi all,

I'm in the process of developing a test strategy for a large bunch of
Visual Basic (VB) programs.  Products like Rational Visual Test seem
to be able to record actions on a GUI for a playback later.

How difficult would it be to use the win32 Python stuff to try
and capture the interactions or instrument VB programs?  I have
access to the VB source too.


Stu





From dominggus at ratelindo.co.id  Mon Aug 30 00:33:23 1999
From: dominggus at ratelindo.co.id (Dominggus Andy. A)
Date: Mon, 30 Aug 1999 11:33:23 +0700
Subject: (no subject)
Message-ID: <B1E0C61741C3D111A8E7006097C3B5A403BAE633@wbmail.ratelindo.co.id>




From bart.lateur at skynet.be  Sat Aug 14 03:52:51 1999
From: bart.lateur at skynet.be (Bart Lateur)
Date: Sat, 14 Aug 1999 07:52:51 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cec7@cs.colorado.edu>
Message-ID: <37b6185f.2442954@news.skynet.be>

John W. Stevens wrote:

>By your reasoning, Perl is very, very inconsistent (since not every
>data type supports '+').

You're looking for the Holy Grail of orthogonality, again. Python is
designed with orthogonality in mind, while Larry Wall allegedly once
said:

    "Any computer scientist who praises orthogonality should be
     sentenced to use an Etch-a-Sketch."

(borrowed from a TomC sig)

	Bart.



From vitus at wagner.rinet.ru  Tue Aug 31 14:40:52 1999
From: vitus at wagner.rinet.ru (Victor Wagner)
Date: 31 Aug 1999 22:40:52 +0400
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <37c2a462.349531009@192.77.87.74> <935523275snz@vision25.demon.co.uk>
Message-ID: <7qh7jk$gaa$1@wagner.wagner.home>

Phil Hunt <philh at vision25.demon.co.uk> wrote:


: The Parrot tool has other potential advantages, e.g: 

: 1. you only have to learn one syntax, that can then be used for
: lots of languages and GUI libraries.

Which is completely useless, becouse you have to learn new syntax anyway
if you want your program do something more than draw a pictures on
screen. You need to learn details of language and GUI library to take
advantage of it. If you don't - why change a library? And this is main
reason of obscure errors, inefficient code and other bad things which
happen to frequently - attempt to use tool without learning it. It is
better to use only one language and one toolkit, but know it well.

: 2. assuming the right import and export tools have been written,
: Parrot can allow you to use one GUI builder to output to a different
: GUI, for example you could design your GUI in Glade, and use Parrot
: to output it as a series of C++ files for KDE/Qt.

I've assumed that whole idea behind Parrot is that GUI builders are
evil, which I agree wholehearthly, becouse there is (yet) no way to
create functionality using point'n'click appoach and functionality is
the first thing programmer should care about.


-- 
--------------------------------------------------------
Victor Wagner @ home       =         vitus at wagner.rinet.ru 
I don't answer questions by private E-Mail from this address.



From mlh at idt.ntnu.no  Wed Aug 11 09:58:28 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 11 Aug 1999 15:58:28 +0200
Subject: report generation
References: <m11Dw1o-0006P6C@iflg>
Message-ID: <y0jiu6mh117.fsf@drue.idi.ntnu.no>

jsouto at dnet.br (Jairo Souto) writes:

> I couldn't find a module for [plain-text] report generation
> from generic data (sql data-base, lists, etc..).
> 
> Is there such a module?

Could you be more specific as to what you wish to accomplish? Pythons
built-in string formatting abilities make this sort of thing quite
easy in general - at least that is my experience (i.e. no need for
separate module.)

> 
> --Jairo Souto  jsouto at dnet.br  (+55) (38) 9971-0014
> 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From greg.ewing at compaq.com  Fri Aug 27 00:31:00 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 27 Aug 1999 16:31:00 +1200
Subject: Extending Python: exception handling
References: <3.0.6.32.19990824122931.0099e860@mail.triode.net.au>
	 <37bdcb30.796706964@news.triode.net.au>
	 <14271.18469.757616.424717@dolphin.mojam.com>
	 <3.0.6.32.19990824122931.0099e860@mail.triode.net.au> <3.0.6.32.19990825123152.009a9100@mail.triode.net.au>
Message-ID: <37C61484.9759CFD4@compaq.com>

John Skaller wrote:
> 
> The types of objects are NOT defined by Viper.
> They're defined in Python by a client module!
> So you can, for example, define any methods you like
> for any type. Here is the current definition of
> the type of an integer in the module 'py_types':
> 
>         class PyIntType:
>                 def succ(x): return x + 1
>                 def pred(x): return x - 1

That sounds very interesting! I've been pondering
the problem of type/class unification in Python,
and your solution is an intriguing one. But is
it possible to do this:

  class MyInteger(PyIntType):
    ... customised behaviour ...

and have it inherit all the default behaviour
of built-in integers? If so, how exactly does it
work? For instance, can I do

  x = MyInteger(3)
  y = x + 5

and, assuming I haven't given MyInteger a custom
__add__ method, have it somehow invoke the built-in
integer addition code.

Greg



From lr at hpl.hp.com  Sun Aug 29 01:39:18 1999
From: lr at hpl.hp.com (Larry Rosler)
Date: Sat, 28 Aug 1999 22:39:18 -0700
Subject: Larry's and Guido's REAL sigs
References: <7q9v2u$1jk$1@nnrp1.deja.com>
Message-ID: <MPG.1232675d6e885633989ec9@nntp.hpl.hp.com>

In article <7q9v2u$1jk$1 at nnrp1.deja.com> on Sun, 29 Aug 1999 00:32:30 
GMT, aaron_watters at my-deja.com <aaron_watters at my-deja.com> says...
...
>    http://www.chordate.com/sigs.GIF
> 
> enjoy!  -- Aaron Watters

Can't.  No such page.  Even with lower-case .gif.  Try again?

-- 
(Just Another Larry) Rosler
Hewlett-Packard Laboratories
http://www.hpl.hp.com/personal/Larry_Rosler/
lr at hpl.hp.com



From I.Clarke at strs.co.uk  Mon Aug 16 09:25:09 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Mon, 16 Aug 1999 14:25:09 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com> <37B7E1AC.FF4FF2CF@strs.co.uk> <ylso5kqa58.fsf@windlord.stanford.edu>
Message-ID: <37B81135.1B598171@strs.co.uk>

My apologies, this is wandering off-topic, I will keep it short.

Ok, so your point is that I was inciting flames, but that is not in
itself a bad thing.  I can live with that.

> comp.lang.perl.misc is primarily for talking about Perl, not trying to
> defend Perl to a Python advocate.  I'm sure the converse is true of
> comp.lang.python.  Someone who starts a language advocacy flamewar and
> then plays innocent about it isn't exactly holding the moral high ground.

I never claimed to be perfect, Tom's post just angered me as he was
accusing me of fanning the flames where as in he was contributing much
more to the whole thing than I was.

Ian.



From bwarsaw at cnri.reston.va.us  Thu Aug 19 15:52:48 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 19:52:48 GMT
Subject: More random python observations from a perl programmer 
References: <37BC6986.A66B044E@starmedia.net>
	<199908191634.KAA01333@jhereg.perl.com>
	<14268.21198.934117.241668@anthem.cnri.reston.va.us>
	<14268.23007.239507.166209@weyr.cnri.reston.va.us>
Message-ID: <14268.24720.827523.852568@anthem.cnri.reston.va.us>

    Fred>   Bucks?  I'm supposed to be getting bucks for this?

Oops, sorry.  Don't tell Fred we're all getting rich here.
    
    Fred> Check the FTP site, Barry; GNU info is there thanks to the
    Fred> active Python community.

Ah, that'll teach me to underestimate the Pythoneers!

    Barry> Tom, but I'm sure Fred would gladly take a contribution
    Barry> that generated manpage format from the latex source.  You
    Barry> might even be surprised at the languages he'd accept such a
    Barry> contribution in! :)

    Fred>   Hey, you're making me look bad!  ;-)

Just trying to fire up the masses.  Look what it did for mailing list
managers :)

-Barry



From jeremy at exit109.com  Mon Aug 23 11:03:18 1999
From: jeremy at exit109.com (Jeremy)
Date: Mon, 23 Aug 1999 15:03:18 GMT
Subject: pyKDE
Message-ID: <rs2olmiv5u288@news.supernews.com>

Is anyone using pyKDE?  I'm trying to get it to work.  It compiled fine,
but actually trying to use it, even just 'import kde', results in:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/kde.py", line 7, in ?
    import libkdec
ImportError: ld.so.1: python: fatal: relocation error: file
 /usr/local/lib/python1.5/lib-dynload/libkdecmodule.so: symbol
 className__C10QImageDrag: referenced symbol not found

Any ideas?  This is pyKDE 0.8, Python 1.5.2, KDE 1.1.1, QT 1.44,
Solaris 7.

-- 
Jeremy  |  jeremy at exit109.com
"How extravagant you are, throwing away women like that.  Someday they
 may be scarce."                                   (Casablanca)



From wware-nospam at world.std.com  Wed Aug  4 12:37:39 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Wed, 4 Aug 1999 16:37:39 GMT
Subject: Python microthreads
References: <FFropB.7tp@world.std.com> <933672146.479670@news.adtranzsig.de> <FFw8Bw.CI4@world.std.com> <933751888.232337@news.adtranzsig.de>
Message-ID: <FFy8ur.LHC@world.std.com>

Will Ware schrieb in Nachricht ...
> I haven't looked at Stackless Python, but it sounds like
> an unrelated thing. As I recall, Stackless Python uses the
> C stack for everything...

Dirk-Ulrich Heise (hei at adtranzsig.de) wrote:
> No, Christian wanted to eliminate the use of the C stack
> and works with the frame objects as well. Maybe one of you
> two re-invented the wheel of the other?

It looks like wheel reinvention is definitely happening.
I've now looked a little bit at Christian's work, and it
looks like there is a lot in common. It probably makes
sense, at least in the long term, to implement microthreads
using Stackless Python. I'll need to study his work more
carefully before I'm ready to migrate.

> Anyway, i just wanted to know whether the two things are
> of independent origins...

Yes, they're independent. Christian has already been in
consultation with the gods of Python, so I think his work is
more in the mainstream spirit of the Python codebase. I
don't think CPython loses anything by being stackless, so
maybe stacklessness could be folded into the core release.
So implementing microthreads separately will become an
increasingly bad idea.

I don't know if Christian's version of eval_code2 can be
stepped a controlled number of opcodes. That would need to
be done for microthreads.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From tim_one at email.msn.com  Tue Aug 17 03:47:08 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 03:47:08 -0400
Subject: C's isprint() concept?
In-Reply-To: <37b8def1.474083836@news.triode.net.au>
Message-ID: <000301bee884$b52e7380$112d2399@tim>

[Tim]
> Take a look at string.translate.

[John (Max) Skaller]
> 	And avoid it, since it isn't compatible with unicode or UTF-8
> encodings :-(

I guess Aahz had better avoid len() then too ...

something-that-views-the-world-as-a-raw-bytestream-can't-really-be-
    incompatible-with-anything-ly y'rs  - tim





From bowman at montana.com  Tue Aug 31 09:41:52 1999
From: bowman at montana.com (bowman)
Date: Tue, 31 Aug 1999 07:41:52 -0600
Subject: calling python interpreter
References: <37cbccd8.5014784@news.tiscalinet.it>
Message-ID: <37CBDBA0.E3ED7A41@montana.com>

Kranio wrote:
> 

> but it gave me "access denied" message. The interperter is in the
> right dir and I am the owner of the file; tha same happens being root.
> What is the problem?

'chmod 777 pippo.py' will probably solve it, if a little overkill. You
need execute permissions set, too. do a 'ls -l pippo.py' and see if
there are any 'x'
or just -rw-rw-rw



From dalke at bioreason.com  Thu Aug  5 18:26:34 1999
From: dalke at bioreason.com (Andrew Dalke)
Date: Thu, 05 Aug 1999 16:26:34 -0600
Subject: SGI C++ compile errors
References: <37A88BD7.3723671D@psych.ucsb.edu>
Message-ID: <37AA0F9A.19F0D6B@bioreason.com>

Jerry Tietz <tietz at psych.ucsb.edu> said:
> I am trying to compile an SGI C++ source
> file which contains the call Py_Initializ() but I
> get the following error messages:
> 
> "/usr/people/beall/dev/Python-1.2/Include/mymalloc.h", line 64:

Whew!  That's really Python 1.2?  Can you use the current
version of Python, which is 1.5.2?

It looks like you're hitting something related to the
Great Renaming.  It's before my time, but as I understand
it, things like a Python-defined "malloc" have since been
renamed to Py_Malloc to fix this problem.

My shot in the dark guess is to comment out those two lines
and force casting if the compiler gives warnings elsewhere.

But you really should upgrade to the latest Python.

						Andrew Dalke
						dalke at bioreason.com



From jstevens at bamboo.verinet.com  Sat Aug 14 11:10:36 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: Sat, 14 Aug 1999 15:10:36 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <slrn7rb1nc.cf9.jstevens@bamboo.verinet.com>

On 14 Aug 1999 04:08:05 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>That could simply have been a reference. Or a symbolic reference.
>
>What is fundamental is that a @ tacked on the front indicates that it is an
>array.

:-)

What is so amusing about that, is that you can say that with a straight
face!

:-)

>So given @$fred, even with no knowledge of what that exactly means
>you should be able to tell that it is somehow treating $fred as an array.

No, what any reasonable person would do would be to grab for his
Perl book. . .

>>Yes. . . is it a hash, or a scalar?  If it is a scalar, why
>>is it called dict?  If it is a hash, then why is it prefixed
>>by $?  If this is a reference instead of a scalar, then why
>>doesn't it have it's own special prefix character.  ;->
>
>It's a scalar. It is named dict because TomC called it that.

Yes.  My point exactly.

>It is
>also named that since it is a reference to a hash. I use code like this
>in C quite a bit :

A reference to a hash. . . and yet TC claims that Perl is open to
non-computer scientists.

:-)

Doesn't *ANYBODY* else see the irony in that?

>If you know what it means then why do you continually get it wrong
>throughout this thread?

I don't suppose that you realize that getting wrong simply
proves (and illustrates) my point?

I learned it.  I used it.  I haven't written a new Perl program
in three months.

I come back to it, I get it wrong. . . do you see, yet,
or do you just not get it?

>>You cannot write Perl that resemble Python.  You are required to 
>>use curly braces as block delimiters.
>
>In fact you are not. You are wrong yet again.

Actually, I'm right, and you go on to prove I'm right.

>Here is some code from Damian Conway from the 'Impythonating PERL' thread
>in march.
>
>package impythonate;
>use Text::Tabs;
>my ($active, @bracket) = (0, ('{', ';', '}') );
>sub import
>{
 ^ Look closely. . . see that curly brace?

>And here is his sample code that is now valid perl (although anyone who uses
>it for real code should be killed) :

To late.  You already used a curly brace.  Disproving your point,
in case you hadn't realized it.

John S.



From jeremy at cnri.reston.va.us  Thu Aug 19 13:59:52 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 19 Aug 1999 13:59:52 -0400 (EDT)
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc313b@cs.colorado.edu>
References: <37bc1339@cs.colorado.edu>
	<19990819174800.D5227@xs4all.nl>
	<37bc313b@cs.colorado.edu>
Message-ID: <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us>

>>>>> "TC" == Tom Christiansen <tchrist at mox.perl.com> writes:

  TC> :There are 'raw' strings though, 'r"<string>"', that dont do
  TC> backslashes.

  TC> Those aren't the book, and there are no manpages.  Therefore, in
  TC> many ways, they don't really count.  I know it hurts to hear
  TC> this, and you're all going to jump on me, but please please
  TC> think about it for a bit before you do so.

[I know I said I wasn't going to get into this, but...] I think this
statement is just plain silly.  I use lots of tools, languages,
libraries, etc. that have non-existent or bad manpages.  The raw
string exist regardless of whether there is a manpage that describes
them.  Which is not to say that I'm opposed to manpages; it would
probably be helpful to provide them.

I did think about this before I jumped, but it still seems fair to
jump.  As a possible counterexample to the "if it don't have manpages
it don't exist" argument, I would suggest the C programming language.
I don't believe there are manpages that describe the language itself,
yet I found it (relatively) easy to learn.  The K&R C book is mighty
fine. 

It's also worth noting that the old Lutz book is a little like the
pre-ANSI K&R C book.  You're not going to find ANSI changes in the old 
K&R, and you're not going to find Python 1.5 info in the old Lutz
book.  In neither case does that mean the improvements don't exist.

Jeremy




From gmcm at hypernet.com  Sat Aug 28 00:05:50 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 27 Aug 1999 23:05:50 -0500
Subject: Slow python on particular computer
In-Reply-To: <37C6C982.35F2B199@prescod.net>
Message-ID: <1276313212-39052155@hypernet.com>

Paul Prescod wrote:

> Has anyone seen a case where Python runs REALLY, painfully slowly on
> an otherwise spritely Windows 95 machine? It takes something like 8
> seconds to launch the interpreted and parse a 10 line program.
> Something must be mis-configured.

Disk fragmentation is the most likely problem. Maybe registry 
pollution, especially if you're using COM.

- Gordon



From mlh at idt.ntnu.no  Sat Aug 28 13:31:23 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 28 Aug 1999 19:31:23 +0200
Subject: Fuzzy string matching?
References: <Pine.LNX.4.10.9908281253390.9521-100000@starship.python.net>
Message-ID: <y0jpv07u80k.fsf@vier.idi.ntnu.no>

Moshe Zadka <moshez at server.python.net> writes:

> On 28 Aug 1999, Magnus L. Hetland wrote:
[...]
> > Could perhaps be a bit more readable, though... (Maybe I'll work on
> > that...)
> 
> No need, my good man. We have a saying in Hebrew "The work of the rightous
> gets done by others"

OK - thanks :)

> 
> Added a comment, made a few variables more meaningful, precalculated
> ranges, changed some ; to , and here is the changed, and a bit tested,
> version:

Awright... I guess it's bit more readable now... However, there are a
few things I think I would do differently - surly only a matter of
taste... (and not meant as a sign of ingratitude :)

> 
> def distance(a,b):
>     n, m = map(len, (a,b))

Is the use of map really clearer than not using it here, with only two
variables? (It may be faster since you don't have to look up len
twice, but...)

>     if n>m:  # Make sure n<=m
>         a,b,n,m = b,a,m,n

I guess I wrote this statement in the original - but I think perhaps I
would split the strings and lengths into two different assignments to
make it more readable, really...

>     range_m, range_n= map(range, (m, n))

Hm. Why? Why is range_m better than range(m)?

>     curr = range(0, n+1)
>     for i in range_m:
>         prev, curr = curr, [i+1]
>         for j in range_n:
>             add, dele = prev[j+1]+1, curr[j]+1
>             change = prev[j] + (a[j]!=b[i] and 1 or 0)

Hm. This is perhaps more compact - and the and/or thing *is* an idiom,
but I don't think it's more readable than a plain if-statement...

>             curr.append(min(add, dele, change))
>     return curr[n]
> 

Here's another take - mainly an expression of my tastes in the matter
(for lack of anything more meaningful to do in the evening... ;)


def distance(a,b):
    "Calculates the Levenshtein distance between a and b."
    n, m = len(a), len(b)
    if n > m:
        # Make sure n <= m, to use O(min(n,m)) space
        a,b = b,a
        n,m = m,n
    current = range(0, n+1)
    for i in range(m):
        previous, current = current, [i+1]
        for j in range(n):
            add, delete = previous[j+1]+1, current[j]+1
            change = previous[j]
            if a[j] != b[i]:
                change = change + 1
            current.append(min(add, delete, change))
    return current[n]


Oh well... There's nothing like fiddling with a little piece of code.
Much more fun than writing large programs methinks :)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From clgonsal at keeshah.penguinpowered.com  Fri Aug 27 03:29:22 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Fri, 27 Aug 1999 07:29:22 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net> <37C552DF.9F90BA36@megsinet.net> <slrn7sanj3.n0r.abigail@alexandra.delanet.com>
Message-ID: <slrn7scfi6.pa.clgonsal@keeshah.penguinpowered.com>

On 26 Aug 1999 10:33:03 -0500, Abigail <abigail at delanet.com> wrote:
> David Oppenheimer (davidopp at megsinet.net) wrote on MMCLXXXVI September
> MCMXCIII in <URL:news:37C552DF.9F90BA36 at megsinet.net>:
> $$ 
> $$ That being said, are there ways to integrate Python and Perl to do things
> $$ that neither one can do alone?!
> 
> Since Python and Perl are Turing equivalent, what things are you refering to?

Write a powerful and useful script that is both:
 - easy to understand and maintain
 - only 6 characters long

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From nharlow at my-deja.com  Tue Aug 31 18:30:31 1999
From: nharlow at my-deja.com (nharlow at my-deja.com)
Date: Tue, 31 Aug 1999 22:30:31 GMT
Subject: Generating an event in Tkinter
References: <7qfc5l$mll$1@nnrp1.deja.com> <011701bef37e$bcd9ece0$f29b12c2@secret.pythonware.com>
Message-ID: <7qhl27$cgp$1@nnrp1.deja.com>

In article <011701bef37e$bcd9ece0$f29b12c2 at secret.pythonware.com>,
  "Fredrik Lundh" <fredrik at pythonware.com> wrote:
> nharlow at my-deja.com wrote:
> > I want to generate a dummy event from within a program.  Then bind a
> > callback to this event.
> >
> > How do I do that?
> >
> > In straight tk, it would be something like:
> >  event generate parent_window <<foo>>
>
> parent_window.event_generate("<<foo>>")
>
> (additional options can be given as
> keyword arguments)

I want to do the following:
iterate over a while loop and output some text to a text widget.
sleep for one second between iterations
here's the code I tried.  It doesn't like the way I am trying to make
it sleep.


i=sleeper=0
from time import sleep
name="analyzer"
while i< 100:
	print "%s: This is %s time through the loop"%(name,i+1)
	widget.event_generate("<<foo>>")
	widget.bind("<<foo>>",lambda x=sleep, y=1:x(y))
	i=i+1

I get the following error:
Traceback (innermost last):
  File "c:\program files\python\lib\lib-tk\Tkinter.py", line 752, in
__call__
    return apply(self.func, args)
  File "C:\nick\GUI\spidergui.py", line 145, in <lambda>
    widget.bind("<<foo>>",lambda x=sleep, y=1:x(y))
AttributeError: no __call__ method defined


what is the best way to do what I am trying to do?

Nick


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From guido at cnri.reston.va.us  Thu Aug 19 14:35:15 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 19 Aug 1999 14:35:15 -0400
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <y0jiu6edoa9.fsf@vier.idi.ntnu.no> <xj0d7wl9acd.fsf@cyclone.harvard.edu> <37BAB643.9D4E861E@strs.co.uk>
Message-ID: <5laernty8s.fsf@eric.cnri.reston.va.us>

Ian Clarke <I.Clarke at strs.co.uk> writes:

> I did some timing tests using the code I have attached below, here are
> the results (in seconds) for sorting a list of 200 items 10 times:
> 
> Dan Schmidt's solution (using sort): 14.98
> Stephan Houben's solution          : 2.82
> Skip Montanaro's solution          : 2.736
>  (as modified by Dan schmidt)
> Magnus L Hetland's solution        : 0.032
> Skip's solution but with arrays    : 3.182
> 
> As you can see I thought that using arrays may provide an efficiency
> speed-up for Skip's algorithm, but apparently not.  Oh, I haven't really
> checked this code too carefully, so feel free to redo the tests if you
> find any bugs in the code.
> 
> Magnus L Hetland's solution wins hands down.

The two-orders-of-magnitude-faster result should have made you look
twice.  There's a bug: s4(list) modifies its argument (empyting it) so
you did only one run, plus 9 runs of an empty list!  Corrected, I get
these timings (on a slightly faster machine, it seems):

s1 12.1
s2 2.31
s3 2.63
s4 2.68
s5 2.82

Here's the fixed s4():

def s4(list):
    """Magnus L Hetland's solution"""
    result = []
    list = list[:]
    for i in range(len(list)):
        element = random.choice(list)
        list.remove(element)
        result.append(element)
    return result

But the winner is Tim Peters' entry, already in the CVS tree.  Its
time is 1.34 seconds:

def shuffle(x, random=random, int=int):
    """x, random=random.random -> shuffle list x in place; return None.

    Optional arg random is a 0-argument function returning a random
    float in [0.0, 1.0); by default, the standard random.random.

    Note that for even rather small len(x), the total number of
    permutations of x is larger than the period of most random number
    generators; this implies that "most" permutations of a long
    sequence can never be generated.
    """

    for i in xrange(len(x)-1, 0, -1):
        # pick an element in x[:i+1] with which to exchange x[i]
        j = int(random() * (i+1))
        x[i], x[j] = x[j], x[i]

--Guido van Rossum (home page: http://www.python.org/~guido/)



From dfan at harmonixmusic.com  Thu Aug 19 13:35:44 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 19 Aug 1999 13:35:44 -0400
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7ro9ns.h6v.jblaine@shell2.shore.net> <37bc2ee5@cs.colorado.edu>
Message-ID: <wkzoznu0zz.fsf@turangalila.harmonixmusic.com>

Tom Christiansen <tchrist at mox.perl.com> writes:

| Online searchable documentation is critical in a technical subject
| like this.  Imagine libc without manpages!  Or having no manpages
| for the shell tools. Now, you might now be aware of this, but I
| happened to *write* most of that on-line documenation for Perl.  I'm
| super senstitive to not being able to look up docs using standard
| CLI text-based tools.  As you said, you would be frustrated not to
| have docs for tools or libraries.

I use the documentation in info format, which is basically plaintext
and easily searchable with text-based tools.  Select "GNU Info" from
<http://www.python.org/doc/current/download.html>.

I'm not trying to be an apologist for the lack of man pages, just
helping you find the plaintext documentation you (and I) prefer.

(no courtesy cc's, please)

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From cfelling at iae.nl  Mon Aug 23 18:10:35 1999
From: cfelling at iae.nl (Carel Fellinger)
Date: 24 Aug 1999 00:10:35 +0200
Subject: matching any number of any character
References: <wr34shrtkk1.fsf@gargoyle.cs.uchicago.edu>
Message-ID: <7psgsr$hh$1@vvs.superst.iae.nl>

Lyn A Headley <laheadle at boguscs.uchicago.edu> wrote:

> accomplish such a task.  I thought this would work:

it does...

>>>> import regex
>>>> rx = regex.compile('\(.\|\n\)+')
>>>> rx.match('abc')
> 3

you see, it just matched 3 characters, but...

>>>  rx.group(1)
> 'c'

here you are tricked by the semantics of a quantified grouping operator.
A quantified group only remembers the last match, and as your group only
matches a single character, 'c' it is:)
(( shouldn't this be documented in the re module? it has stung more people
   recently ))

> but I wanted 'abc'!

so you have to refer to the whole match (rx.group(0)) or get the
quantification inside a group. e.g.

>>> rx = regex.compile('\(\(.\|\n|)+\)')
>>> rx.group(1)
'abc'
>>> rx.group(2)
'c'

-- 
groetjes, carel



From arnold at dstc.edu.au  Mon Aug 23 03:09:30 1999
From: arnold at dstc.edu.au (David Arnold)
Date: Mon, 23 Aug 1999 07:09:30 GMT
Subject: carpe linguam 
In-Reply-To: Your message of "Mon, 23 Aug 1999 02:18:20 +0100."
             <0Hg8yBAcFKw3EwjN@jessikat.demon.co.uk> 
References: <0Hg8yBAcFKw3EwjN@jessikat.demon.co.uk>
Message-ID: <199908230709.RAA03985@piglet.dstc.edu.au>

-->"Robin" == Robin Becker <robin at jessikat.demon.co.uk> writes:

  Robin> Oh dear are there any transputer fans left out there?

yes ... ;-)



d




From andrew-johnson at home.com  Fri Aug 20 03:52:50 1999
From: andrew-johnson at home.com (Andrew Johnson)
Date: Fri, 20 Aug 1999 07:52:50 GMT
Subject: making python man pages looks hard
References: <1103_935134343@shoal>
Message-ID: <mP7v3.6033$dr6.136269@news1.rdc2.on.home.com>

In article <1103_935134343 at shoal>,
 Dan Connolly <connolly at w3.org> wrote:
[snip] 
! parting shot: CPAN is cool, but I find it frustrating that I can't
! read the documentation for a module without downloading
! and upacking the module. for example, I can browse
! the list of modules,
! 
! http://www.perl.com/CPAN/modules/00modlist.long.html
! 
! but say I find one I'm interested in:
! 
! Mac::Comm::
! ::OT_PPP        RdpO Control Open Transport PPP / Remote Access   CNANDOR
! 
! the only link goes to the Author contact info. Gee thanks.

that is usually no longer a valid complaint --- try the CPAN
front end at 'theory.uwinnipeg.ca', in particular, the search
engine:

http://theory.uwinnipeg.ca/search/cpan-search.html

I entered Mac::Com in the box and got back:

Mac-Comm-OT_PPP-1.20.tar.gz [readme / module list] 3.6 1998-01-06 CNANDOR
Mac::Comm::OT_PPP [ documentation]: Interface to Open Transport PPP

(and yes, those are links to the .tgz itself, the readme, a list of
modules in the package; and, surprise, "[documentation]" links to the
html'ified manpage).

regards
andrew

-- 
      They're not soaking, they're rusting!
          -- my wife on my dishwashing habits
      



From philh at vision25.demon.co.uk  Sat Aug 14 06:31:39 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sat, 14 Aug 99 10:31:39 GMT
Subject: Does Python run under Windows 2000? (build 2031)
References: <37B40D58.DD6AADC0@RobotVision.com>
Message-ID: <934626699snz@vision25.demon.co.uk>

In article <37B40D58.DD6AADC0 at RobotVision.com>
           jgreen at RobotVision.com "James E. Green" writes:
> I'm running Windows 2000 RC1 (build 2031, the very latest).  Will Python
> run under this OS?

Use Linux instead. You know it means sense. :-)

-- 
Phil Hunt....philh at vision25.demon.co.uk




From a.eyre at optichrome.com  Tue Aug 10 08:52:48 1999
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 10 Aug 1999 12:52:48 GMT
Subject: __init__ question
In-Reply-To: <1278154513-26981346@hypernet.com>
References: <1278154513-26981346@hypernet.com>
Message-ID: <000901bee32f$3f82fe30$3acbd9c2@peridot.optichrome.com>

> > a = Vector()  # x=0 y=0 z=0
> > a = Vector( 1)  # x=1 y=1 z=1
> > a = Vector( 1, 2, 3)  # x=1 y=2 z=3

>  def __init__(self, x=0, y=0, z=0):

How about:

>>> class Vector:
...  def __init__(self, *args):
...   print args
...
>>> v=Vector()
()
>>> v=Vector(1)
(1,)
>>> v=Vector(1,2)
(1, 2)

...etc.

You can then parse the args in the constructor.

--------------------------------------------
Adrian Eyre <mailto:a.eyre at optichrome.com>
Optichrome Computer Solutions Ltd
Maybury Road, Woking, Surrey, GU21 5HX, UK
Tel: +44 1483 740 233  Fax: +44 1483 760 644
http://www.optichrome.com 
--------------------------------------------




From mwh21 at cam.ac.uk  Tue Aug 24 08:20:27 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 24 Aug 1999 13:20:27 +0100
Subject: Can't use crypt
References: <Pine.LNX.4.10.9908241244050.21229-100000@snet.wit.ie>
Message-ID: <m3n1vh74l0.fsf@atrus.jesus.cam.ac.uk>

"Kieran O' Sullivan" <kosullivan at snet.wit.ie> writes:

> When I tyr to compile python 1.5.2 on SuSE Linux 6.0 with the crypt module
> I get the following error.
> 
> make[1]: Entering directory `/Python-1.5.2/Modules'
> gcc  -Xlinker -export-dynamic python.o \
>           ../libpython1.5.a   -lieee -ldl  -lm  -o python
> ../libpython1.5.a(cryptmodule.o): In function `crypt_crypt':
> /Python-1.5.2/Modules/./cryptmodule.c:21: undefined reference to `crypt'
> collect2: ld returned 1 exit status
> make[1]: *** [link] Error 1
> make[1]: Leaving directory `/Python-1.5.2/Modules'
> make: *** [python] Error 2
> 
> I have compiled python 1.5.2 with crypt on SuSE 5.2 with no problems.
> There was a previous copy of python 1.5.1 on the SuSE 6.0 machine but
> python 1.5.2 cpmpiles without any problem when crypt is not included.

So edit Modules/Setup to add -lcrypt then...

> When I use -lcrypt instead python will compile but I can't use cthe crypt
> function.

Oh. What goes wrong now?

Michael



From mlh at idt.ntnu.no  Fri Aug 20 15:40:34 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 20 Aug 1999 21:40:34 +0200
Subject: More random python observations from a perl programmer
References: <37BC6986.A66B044E@starmedia.net> <199908191634.KAA01333@jhereg.perl.com> <14268.21198.934117.241668@anthem.cnri.reston.va.us> <14268.23007.239507.166209@weyr.cnri.reston.va.us>
Message-ID: <y0j7lmq44wd.fsf@vier.idi.ntnu.no>

"Fred L. Drake, Jr." <fdrake at acm.org> writes:

> Barry A. Warsaw writes:
[...]
>   Manpages have never been requested, unless you consider Tom's
> complaint a request.  I've received nothing at
> python-docs at python.org.  Certainly not a contributed conversion
> script.

Hm. Could I request them now? I think it might be kinda nice to
have... :)

(Though I'm not sure I would want to make a conversion script... Does
there really not exist one, either from tex, dvi or info to groff?
Hm...)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From eye at niksula.hut.fi  Tue Aug 24 07:43:03 1999
From: eye at niksula.hut.fi (Petri Mikael Kuittinen)
Date: 24 Aug 1999 14:43:03 +0300
Subject: How to sell Python to your boss?
References: <11A17AA2B9EAD111BCEA00A0C9B41793034AAD03@molach.origin.ea.com> <14268.19435.271581.609@bitdiddle.cnri.reston.va.us>
Message-ID: <ka8so59jtfc.fsf@cheerful.cs.hut.fi>

Jeremy Hylton <jeremy at cnri.reston.va.us> writes:

> You might also pick up some skills that would help you find a job.

I think learning Python has made me a better C/assembler/Java/whatever
programmer as well. By learning new programminmg languages one usually
is able to pick up new programming methods and techniques, programming
patterns to solve probles etc. Most of these are applicable to other
languages as well.

That is why I have also taken a look at exotic languages such as Ruby,
Lua and Pliant. They all have some good ideas.


Petri

-- 
<(O)> Petri Kuittinen, also known as Eye, Dj Eye or Peku               <(O)>
<(O)> ADDRESS: Timpurinkuja 1 C 39, FIN-02600 Espoo, Finland           <(O)>
<(O)> EMAIL: eye at iki.fi WWW: http://www.iki.fi/~eye/ PHONE: 09-5472380 <(O)>
~Humor is the best antidote to reality.



From ivanlan at callware.com  Thu Aug 19 13:19:55 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Thu, 19 Aug 1999 17:19:55 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
		<slrn7ro8n1.h6v.jblaine@shell2.shore.net> <37bc29d8@cs.colorado.edu>
Message-ID: <37BC3CBB.7D28E3E5@callware.com>

Hi All--

Tom Christiansen wrote:
> 

[bobbit]

> :Thanks for the document.  It was very interesting reading.
> 
> You're welcome.  I got so fed up with the damned flames coming
> from people who don't actually even know what they're talking about,
> that I decided to make sure I wasn't one of them. :-)
> 
> --tom
> 

If that's what you believe, why are posting stuff like this (from a
different, but nearly simultaneous, posting)?

> You seem to be looking for a fight.   Please say nothing
> at all if that's all you want to do.

In response to someone else's quite reasonable suggestion to read the
documentation.

You also say in the same message

> Thanks for nothing: there are no manpages.  I read the Lutz book.
 
It's true that there are no manpages.  Linux comes with manpages, but
every single manpage says (paraphrased): ``Don't trust this page; we
don't.  See the info help system for the real scoop.''  When you do,
it's harder to use and usually doesn't have any more to say than the
manpage.  I'm not sure that the choice of two mostly wrong options is a
choice with any value.

Python doesn't have manpages, true.  It does have perfectly adequate
documentation available at 

	http://www.python.org/doc

You can download that documentation and install it on your own computer
if you wish to avoid the net.  Like all documentation, it leaves
something to be desired (decent search facilities, for one, but manpages
don't have those either); but, like Kernighan & Ritchie's _The C
Programming Language_, it's all there.

Your complaint that Python doesn't have manpages is like complaining
that Buddhism doesn't have a Bible.  Perfectly true, and perfectly
false.

<a-difference-that-doesn't-make-a-difference-isn't-information>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From tchrist at mox.perl.com  Fri Aug 20 09:22:14 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 20 Aug 1999 06:22:14 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <37bc313b@cs.colorado.edu> <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us> <019301beeaea$004c4d80$f29b12c2@secret.pythonware.com>
Message-ID: <37bd4876@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    "Fredrik Lundh" <fredrik at pythonware.com> writes:
:as an aside, it's a bit sad that Tom didn't read the book's
:preface, where Mark says, among other things:

It's a shame that so many of you should be so unbelievably harsh on me
when *I'm* the one who with malice toward none and full of revulsion
at the flames causes by *YOUR* community's ignorant attacks on mine,
that *I* made the honest effort to learn *YOUR* language, and that I
furthermore for your benefit not my own then took the additonal trouble
to write-up my experiences and observations gleaned while five hundred
miles away from the nearest tree or internet link.

And for that, I am abused for my candor and effort in trying to convey
to you what that process was like for me.  

Shame on you!

Don't you understand your gain?  How dare you punch a gifthorse in the
mouth--repeatedly?  If these petty slams and disdainful arrogances are the
kind of response typical of your community, this bodes not well for you.

But no, that's not it.  It's just Usenet, the place where condescension,
conceit, and cruelty all come far too easily.  The more the pissier,
you know.  I'll try to keep that in mind as I continue to attempt to
sift through this dreck to find the gems.

--tom
-- 
    "I will always write it the second way, so if you're optimizing for
    contrariness, it's obviously better to do it the second way."
    	--Larry Wall



From cjc26 at nospam.cornell.edu  Sat Aug 21 15:15:24 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 21 Aug 1999 19:15:24 GMT
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <Pine.LNX.4.10.9908200706500.15847-100000@starship.python.net> <37bdc47f.22030506@news.erols.com>
Message-ID: <slrn7rtgod.9l.cjc26@tankgrrl.bridget.mindriot.net>

Pada Fri, 20 Aug 1999 21:12:55 GMT, Robert Kern bilang:
| >> > Funny, I read that 'the way g-d indented'. :)
| >> 
| >> Hey, what a great signature that would make!
| >> 
| >> Python: Programming the way g-d indented.
| >
| >Make that:
| >Python: Programming the way g--d- indented.
| 
| Why does this remind me of Douglas Hofstadter's _G?del, Escher, Bach_?
| :-)

Programming the way g-d-- intended?  Now that's just scary..


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From aahz at netcom.com  Mon Aug 16 19:53:45 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Aug 1999 23:53:45 GMT
Subject: C's isprint() concept?
References: <000201bee78f$6bbf6ec0$f22d2399@tim> <7p9d0v$m9c@dfw-ixnews11.ix.netcom.com> <37B8A265.8A9EAC69@vision.arc.nasa.gov>
Message-ID: <7pa8a9$pik@dfw-ixnews8.ix.netcom.com>

In article <37B8A265.8A9EAC69 at vision.arc.nasa.gov>,
Chad Netzer  <chad at vision.arc.nasa.gov> wrote:
>Aahz Maruch wrote:
>>
>> Ah.  Interesting.  I'm using re.sub because we're still on 1.5.1, but
>> we're supposed to move to 1.5.2 Real Soon Now.
>
>Just out of curiosity, what is the biggest source of the delay from
>switching to 1.5.2?  Is it a matter of "if it ain't broke...", or do the
>changes warrant enough work (rewriting code, updatin libraries,
>etc.) that it is a big job?  I'm writing apps, and currently rely on
>some 1.5.2 behaviors, so I'm curious as to how difficult it may be
>for customers to upgrade if need be.

There were a couple of specific bugs in 1.5.1 that required us to create
a special Python package, and it will be a little tricky to back out
those changes (I wasn't involved, so don't ask for more details).  Plus
we simply haven't had the time to do the necessary regression testing.

(We've got Zope, Medusa, mxODBC, and a slew of other bits; just the back
end currently has, um, about twelve running processes.)

Overall, if your customers are running unpatched Python distributions
and have not written workarounds for bugs in 1.5.1, upgrading should be
a snap.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From forcer at mindless.com  Mon Aug 23 17:32:44 1999
From: forcer at mindless.com (forcer)
Date: 23 Aug 1999 23:32:44 +0200
Subject: Necessity of ``pass''
Message-ID: <hhemgu41z7.fsf@forcix.roof.lan>

Hi there.

Lately, i've been thinking about Python. While it is an
exceptionally clean language, it has one very obvious wart:
The ``pass'' statement.

To quote the reference manual:

   pass is a null operation -- when it is executed, nothing
   happens. It is useful as a placeholder when a statement is
   required syntactically, but no code needs to be executed, for
   example:

   def f(arg): pass    # a function that does nothing (yet)

So it happens that Python has places where semantically nothing
should be done, while syntactically it requires a statement.

I consider this suboptimal.

As a solution, one could allow the following:
instead of

	if foo:
	   pass
	bar

one could write

	if foo:
	bar

Syntactically this does not introduce any ambiguity. The only
problem i see is that the "pass" version is more readable.

But...
Most of the time, people write

        if foo:pass
        bar

anyways so the "readability" does not suffer from leaving pass
out.

Also, the requirement for pass makes it impractical to comment
out parts of an if expression, like in

        if foo:
	   # baz
	   # qux
	bar

The new syntax would even allow a smooth transition, or
coexistence of both versions. Pass can be retained to keep the
possibility for a explicit do-nothing statment, while it's just
not mandatory anymore.

I'd like to see such a change in the Python language, so it'll be
even cleaner than it was before.

I hope my opinions and thoughts are of value to you,
       --forcer

-- 
((email . "forcer at mindless.com")       (www . "http://forcix.cx/")
 (irc   . "forcer@#StarWars (IRCnet)") (gpg . "/other/forcer.gpg"))



From dwelton at linuxcare.com  Mon Aug  9 21:54:07 1999
From: dwelton at linuxcare.com (David N. Welton)
Date: 09 Aug 1999 18:54:07 -0700
Subject: Redirect stderr
Message-ID: <87lnbk8kow.fsf@linuxcare.com>

I wish to do this:

#!/usr/bin/python

import sys
import os

f = open("goober", "w")

sys.stderr = f
sys.__stderr__ = f

os.execv("/bin/ls", ["/bin/ls", "sdfsdfdsfds"])

and have it dump the errors into 'goober'.  I'm having some
difficulties figuring out exactly why this isn't working, or how I
ought to be doing this.

Thanks,
-- 
David N. Welton, Web Engineer, Linuxcare, Inc.
415.354.4878 x241 tel, 415.701.7457 fax
dwelton at linuxcare.com, http://www.linuxcare.com/
Linuxcare. At the center of Linux.



From mwh21 at cam.ac.uk  Wed Aug 25 07:35:12 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 25 Aug 1999 12:35:12 +0100
Subject: [Python] Can't use crypt on SuSE 6.0
References: <Pine.LNX.4.10.9908251052460.2094-100000@snet.wit.ie>
Message-ID: <m3lnb014b3.fsf@atrus.jesus.cam.ac.uk>

Oops, I meant to follow up to your follow up to my follow up to your
original post... but I never got round to it.

"Kieran O' Sullivan" <kosullivan at snet.wit.ie> writes:
> When I tyr to compile python 1.5.2 on SuSE Linux 6.0 with the crypt module
> I get the following error.
>  
> make[1]: Entering directory `/Python-1.5.2/Modules'
> gcc  -Xlinker -export-dynamic python.o \
>           ../libpython1.5.a   -lieee -ldl  -lm  -o python
> ../libpython1.5.a(cryptmodule.o): In function `crypt_crypt':
> /Python-1.5.2/Modules/./cryptmodule.c:21: undefined reference to `crypt'
> collect2: ld returned 1 exit status
> make[1]: *** [link] Error 1
> make[1]: Leaving directory `/Python-1.5.2/Modules'
> make: *** [python] Error 2
> 
> I have compiled python 1.5.2 with crypt on SuSE 5.2 with no problems.
> There was a previous copy of python 1.5.1 on the SuSE 6.0 machine but
> python 1.5.2 cpmpiles without any problem when crypt is not included.

That'll be a libc5 vs. libc6 issue, I'd warrant. This is not mysterious.

> When I use -lcrypt option in the Setup file python will compile but I
> can't use the crypt function.
> 
> When I try I get ImportError: no module named crypt

*This* is the odd bit. Are you sure you're executing the right Python?
I've done things like this to myself where I compile & install a
package in /usr/local and then carry on getting the version that's
installed in /usr...

> I have tried the following command with crypt inabled:
> 
>   strace -o /tmp/foo -f make install 
> 
> but the file that this produced is not a lot of help. 

I'm curious - how did you expect that to help?

More productive might be the output of

python -vc 'import crypt'

and

python -c 'import sys; print sys.builtin_module_names'

the-link-error-is-not-the-problem-ly y'rs - Michael



From fleck at informatik.uni-bonn.de  Wed Aug 25 08:48:40 1999
From: fleck at informatik.uni-bonn.de (Markus Fleck)
Date: 25 Aug 1999 12:48:40 GMT
Subject: equivalent of getch
References: <37C3C97A.59BF653E@acay.com.au>
Message-ID: <7q0on8$h3i@news.rhrz.uni-bonn.de>

Gurvijay Singh Bhatti <khalsa at acay.com.au> wrote:
> could someone tell me what is equivalent of getch in jpython? I want the
> program to wait for the user to hit enter before going forward.

AFAIK, this cannot be done in Java. I would
be surprised if it could be done in JPython.

Yours,
Markus.

-- 
////////////////////////////////////////////////////////////////////////////
   Markus B Fleck - University of Bonn - CS Department IV - WHOIS MF5079
          UNIX Administrator - comp.lang.python.announce Moderator
     "GNU Gather" Free Internet Groupware Project - http://gather.net/
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\



From aarhodes at cisco.com  Tue Aug 10 12:03:43 1999
From: aarhodes at cisco.com (Aaron Rhodes)
Date: Tue, 10 Aug 1999 16:03:43 +0000
Subject: What's a good way to get my IP from Python?
Message-ID: <37B04D5F.2F00EE6C@cisco.com>

Hello,

Does anyone know of a really good way to determine the
IP address of the ethernet cards on a Linux system
using Python only (i.e. no C code/parsing external binaries, etc.) ?

Thanks in advance,

Aaron

--
Aaron Rhodes
Network Security Engineer
aarhodes at cisco.com
(512) 378-1097






From MHammond at skippinet.com.au  Tue Aug  3 19:13:53 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 4 Aug 1999 09:13:53 +1000
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk>
Message-ID: <7o7t4p$rgp$1@m2.c2.telstra-mm.net.au>

Ian Clarke wrote in message <37A6A2DB.9B11FB01 at strs.co.uk>...

>It is to create a Python web server which automatically parses embedded
>Python in HTML (there is a Perl extension that does this too).  So, for
>example, the following HTML code:

Attached is some code I wrote for exactly this purpose.  I looked into Zope
and it's DocTemplate class, but it had some restrictions that I didnt like
for my purpose.  I wanted something closer to what you describe.

I stole my implementation idea from Microsoft's ASP.  <%  and <%= tags
delimit Python code.  There is no requirement on the text outside the <%
tags, so HTML (or indeed many other things) work fine.

<%= indicates a Python expression to be evaluated and the result placed in
the output stream.
<% indicates a Python block to be executed.  A "write" function can be
called by this code to insert text into the output stream.

The application provides a dictionary to this tool - this dictionary is the
"namespace" made available to the executed code.  there is no support for
rexec, as I trust the users running this :-)

A special token <% #end %> is used to overcome Python's white-space in this
environment.  Indentation is automatically applied to subsequent code
blocks.

Example text may be:
<% for i in range(3): %>
<%= "This is loop number %d" % i %>
<% #end %>
<% import string # Keep newline in token so not in output!
%><% words = string.split("hi there") %>
<% for word in words: %><%= word %><% #end %>

Note that the second line of the example is not indented, but magically
appears in the for loop.  The <% #end %> token ends the block, meaning the
import is not indented.

Attached for your amusement.

Mark.



begin 666 doctemplate.py
M:6UP;W)T('-Y<RP@<W1R:6YG+"!T<F%C96)A8VL-"FEM<&]R="!R90T*#0IR
M94)L;V-K4W1A<G0@/2!R92YC;VUP:6QE*'(B+BHZ7',J*",N*BD_)"(I(",@
M36%T8V@@86YY=&AI;F<@96YD:6YG(&EN("(Z(BP at 97AC;'5D:6YG(&-O;6UE
M;G1S+ at T*9&5B=6=,979E;" ](# -"@T*9&5F($AA;F1L945X8V5P=&EO;B at I
M. at T*"70L('8L('1B(#T@<WES+F5X8U]I;F9O*"D-"@ER971U<FX@(B-%<G(@
M)7,Z("5S(B E("AT+"!V*0T*#0ID968 at 0VAU;FM&:6QE*&8I. at T*"6-H=6YK
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M("(@*B!I;F1E;G1,979E; T*"0EI9B!S=&%R=%!O<SYL87-T4&]S. at T*"0D)
M8VAU;FMS+F%P<&5N9"AI;F1E;G0@*R G=W)I=&4H)7,I(V=E;EQN)R E(&!D
M871A6VQA<W10;W,Z<W1A<G10;W-=8"D-"@D)(R!7;W)K(&]U="!W:&%T('1Y
M<&4@;V8 at 8VAU;FL@=&AE(&YE>'0@;VYE(&ES#0HC"0EI9" ](&1A=&%;<W1A
M<G10;W,K,CIS=&%R=%!O<RLS70T*"0EC;V1E(#T@<W1R:6YG+G-T<FEP*&1A
M=&%;<W1A<G10;W,K,CIE;F10;W-=*0T*"0EI9B!C;V1E6S!=(#T]("<])SH-
M"@D)"6-O9&4@/2!C;V1E6S$Z70T*"0D)8VAU;FMS+F%P<&5N9"AI;F1E;G0@
M*R G=W)I=&4H7U]E=F%L*"5S+"!G;&]B86QS*"DL(&QO8V%L<R at I*2E<;B<@
M)2!@8V]D96 I#0H)"65L<V4Z#0H)"0DC($-O=6QD(&)E(&%R8FET87)Y(&-H
M=6YK+"!O<B!S<&5C:6%L(&)L;V-K(&UA<FME<BX-"@D)"6EF(')E0FQO8VM3
M=&%R="YM871C:"AC;V1E+" Q*2!I<R!N;W0 at 3F]N93H-"@D)"0EI9B!S=')I
M;F<N;'-T<FEP*&-O9&4I6SHT72!N;W0@:6X at 6R=E;'-E)RPG96QI9B==. at T*
M"0D)"0DC($ET(&ES(&$@8FQO8VL@<W1A<G0-"@D)"0D)8VAU;FMS+F%P<&5N
M9"AI;F1E;G0@*R!C;V1E("L@(EQN(BD-"@D)"0D):6YD96YT3&5V96P@/2!I
M;F1E;G1,979E;" K(#$-"@D)"0EE;'-E. at T*"0D)"0EI;F1E;G0@/2 B(" @
M("(@*B H:6YD96YT3&5V96PM,2D-"@D)"0D)8VAU;FMS+F%P<&5N9"AI;F1E
M;G0@*R!C;V1E("L@(EQN(BD-"@D)"65L<V4Z#0H)"0D):68@<W1R:6YG+FQS
M=')I<"AC;V1E*5LZ-%T]/2(C96YD(CH-"@D)"0D):6YD96YT3&5V96P@/2!I
M;F1E;G1,979E;" M(#$-"@D)"0D):6YD96YT(#T@(B @(" B("H@:6YD96YT
M3&5V96P-"@D)"0EC:'5N:W,N87!P96YD*&EN9&5N=" K(&-O9&4@*R B7&XB
M*0T*"0D)"0T*"0EL87-T4&]S(#T at 96YD4&]S*S(-"@D)<W1A<G10;W,@/2!S
M=')I;F<N9FEN9"AD871A+" B/"4B+"!L87-T4&]S*0T*"2, at 06YD(&5V97)Y
M=&AI;F<@=&\@=&AE(&5N9"!I<R!L:71E<F%L#0H):6YD96YT(#T@(B @(" B
M("H@:6YD96YT3&5V96P-"@EC:'5N:W,N87!P96YD*&EN9&5N=" K("=W<FET
M92 at E<RDC9V5N7&XG("4 at 8&1A=&%;;&%S=%!O<SI=8"D-"@ER971U<FX at 8VAU
M;FMS#0H-"F1E9B!?7V5V86PH(&5X<'(L(&<L(&PI. at T*"71R>3H-"@D)<F5T
M(#T at 979A;"AE>'!R+"!G+"!L*0T*"0ER971U<FX@<W1R*')E="D-"@EE>&-E
M<'0Z#0H)"70L('8L('1B(#T@<WES+F5X8U]I;F9O*"D-"@D)<')I;G0@(D5X
M<')E<W-I;VX@)R5S)R!F86EL961<;B E<SH@)7,B("4@*'-T<FEN9RYS=')I
M<"AE>'!R*2P@="P@=BD-"@D):68 at 9&5B=6=,979E;#H-"@D)"71R86-E8F%C
M:RYP<FEN=%]E>&,H*0T*"0EI9B!T(&ES($%T=')I8G5T945R<F]R.B C(%1R
M>2!A;F0 at 9VEV92!T:&4@=7-E<B!M;W)E(&AI;G1S+ at T*"0D):71E;7,@/2!S
M=')I;F<N<W!L:70H97AP<BP@(BXB*0T*"0D)=')Y. at T*"0D)"69O<B!I;F1E
M>"!I;B!R86YG92AL96XH:71E;7,I*3H-"@D)"0D);F%M92 ]('-T<FEN9RYJ
M;VEN*&ET96US6SII;F1E>"LQ72PB+B(I#0H)"0D)"79A;" ](&!E=F%L*&YA
M;64L(&<L(&PI8 T*"0D)"0EP<FEN=" B(" E<STE<R(@)2 H;F%M92P@=F%L
M*0T*"0D)97AC97!T. at T*(PD)"0ET<F%C96)A8VLN<')I;G1?97AC*"D-"@D)
M"0EP87-S#0H)"7)E='5R;B B(@T*#0ID968 at 1'5M<$-O9&4H<V]U<F-E+"!S
M=&%R=#TQ+"!E;F0].3DY*3H-"@EP<FEN=" B/2(@*B U, T*"6QI;F5N;R ]
M(# -"@EF;W(@;&EN92!I;B!S=')I;F<N<W!L:70H<V]U<F-E+")<;B(I. at T*
M"0EL:6YE;F\@/2!L:6YE;F\@*R Q#0H)"6EF(&QI;F5N;SQS=&%R=#H-"@D)
M"6-O;G1I;G5E#0H)"6EF(&QI;F5N;SYE;F0Z#0H)"0EB<F5A:PT*"0EI9B!L
M:6YE6RTT.ET]/2<C9V5N)SH-"@D)"6QO;VL](G=R:71E*"(-"@D)"6EN9&5X
M(#T@<W1R:6YG+FEN9&5X*&QI;F4L(&QO;VLI*VQE;BAL;V]K*0T*"0D)=&5X
M=" ](&QI;F5;:6YD97 at Z+35=#0H)"0EI9B!L96XH=&5X="D@/B U-3H-"@D)
M"0ET97AT(#T@=&5X=%LZ,C5=("L@(BXN+B(@*R!T97AT6RTR-3I=#0H)"0EL
M:6YE(#T@;&EN95LZ:6YD97A=("L@=&5X=" K("(I(@T*"0EP<FEN=" B)3-D
M/B5S(B E("AL:6YE;F\L(&QI;F4I#0H)<')I;G0@(CTB("H at -3 -"@T*9&5F
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M9&5B=6=,979E;#XQ. at T*"0E$=6UP0V]D92AS;W5R8V4I#0H)=')Y. at T*"0EC
M;V1E(#T at 8V]M<&EL92AS;W5R8V4L("(\<V-R:7!T/B(L(")E>&5C(BD-"@EE
M>&-E<'0 at 4WEN=&%X17)R;W(L(&1E=&%I;',Z#0H)"6UE<W-A9V4L("AF:6QE
M;F%M92P@;&EN96YO+"!O9F9S970L('1E>'0I(#T at 9&5T86EL<PT*"0E$=6UP
M0V]D92AS;W5R8V4L(&QI;F5N;RTU+"!L:6YE;F\K-2D-"@D)<')I;G0@(DQI
M;F4B+"!L:6YE;F\L(BP at 0V]L(BP@;V9F<V5T+" B.B(L(&UE<W-A9V4-"@D)
M<F5T=7)N#0H)=')Y. at T*"0EE>&5C(&-O9&4@:6X@;F%M97-P86-E#0H)97AC
M97!T. at T*"0EI9B!S>7,N97AC7W1Y<&4@:6X@<F5R86ES86)L945X8V5P=&EO
M;G,Z#0H)"0ER86ES90T*"0EI9B!D96)U9TQE=F5L. at T*"0D)=')A8V5B86-K
M+G!R:6YT7V5X8R at I#0H)"7!R:6YT(")%>&5C=71I;VX at 9F%I;&5D(@T*"0EP
M<FEN=" B)7,Z("5S(B E("AS>7,N97AC7VEN9F\H*5LP72P@<WES+F5X8U]I
M;F9O*"E;,5TI#0H-"@T*=&5S=" ]("(B(EP-"CPE(&9O<B!I(&EN(')A;F=E
M*#,I.B E/CPE<')I;G0@:24^/"4C96YD)3X-"DQI=&5R86P\)2!I;7!O<G0@
M<W1R:6YG( T*)3X\)3T at 9F]O( T*)3X\)2!W;W)D<R ]('-T<FEN9RYS<&QI
M="@B:&D@=&AE<F4B*2 -"B4^/"4 at 9F]R('=O<F0@:6X@=V]R9',Z("4^/"4]
M('=O<F0@#0HE/CPE("-E;F0@#0HE/DQA<W0 at 3&ET97)A; T*/"4@:68 at 9FQA
M9SH@)3Y)=',@=')U93PE(&5L<V4Z("4^271S(&9A;'-E/"4C96YD)3X-"B(B
M(@T*:68 at 7U]N86UE7U\]/2=?7VUA:6Y?7R<Z#0H)9&5B=6=,979E;#TR#0H)
M:6UP;W)T(&-3=')I;F=)3PT*"61I8W0@/2![?0T*"61I8W1;)V9O;R==(#T@
M)V9O;R!W87,@:&5R92<-"@ED:6-T6R=F;&%G)UT@/2 Q#0H)9&EC=%LG=W)I
M=&4G72 ]('-Y<RYS=&1O=70N=W)I=&4-"@E%>&5C=71E5&5M<&QA=&4H8U-T
><FEN9TE/+E-T<FEN9TE/*'1E<W0I+"!D:6-T*0T*
`
end




From andrew at starmedia.net  Tue Aug 17 14:21:47 1999
From: andrew at starmedia.net (Andrew Csillag)
Date: Tue, 17 Aug 1999 18:21:47 GMT
Subject: Getting the module name from a method...
References: <7pboln$n74$1@feed.teaser.fr>
Message-ID: <37B9A83B.A987188F@starmedia.net>

Olivier Deckmyn wrote:
> 
> Hi !
> 
> I would like a method of my own to be able to print the module it is
> declared in:
> 
> ex:
> the text for the module : toto.py is :
> 
> class MyClass:
>     def myMethod(self):
>         print "I am in module", ????????
> 
> and then
> MyClass().myMethod()
> should produce :
> 'toto'
> 
> ....
> Thanx for your help !
> 
> ---
> Olivier Deckmyn, Paris - France
> "Any sufficiently advanced technology is indistinguishable from
> magic." -Arthur C. Clark
the name of the current module is __name__
so if you have a module foo.py
def f():
    print __name__

and the main executable:
import foo
foo.f()

will print 'foo'

-- 
"Programmers are just machines that convert coffee to executable code"
   -Unknown

"Drew Csillag" <drew_csillag at geocities.com>



From hnowak at cuci.nl  Sat Aug 28 10:36:22 1999
From: hnowak at cuci.nl (Hans Nowak)
Date: Sat, 28 Aug 1999 14:36:22 GMT
Subject: Fuzzy string matching (snippet candidate?)
In-Reply-To: <37C718AA.544E3FB8@msfc.nasa.gov>
References: <37C718AA.544E3FB8@msfc.nasa.gov>
Message-ID: <199908281437.QAA22455@dionysus.fw.cuci.nl>

On 27 Aug 99, at 23:00, Drew McDowell wrote:

> Magnus,
>   Did you think of submitting this to the python snippets repository?
> http://www.hvision.nl/~ivnowa/snippets/

He doesn't have to; I already added this to the pile of snippets 
sitting on my hard disk. As a matter of fact, the snippet maintainer 
(me ;-) also the one who asked the initial question for the fuzzy 
string match.

Apropos snippets; I haven't been able to work on it for the last 
weeks since I've been on a little vacation. ^_^ I will try to add 
some new snippets this weekend. (Apparently people are actually 
*using* this page. Amazing. ;-)

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/



From tim_one at email.msn.com  Fri Aug 20 01:48:14 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 20 Aug 1999 01:48:14 -0400
Subject: sliciing dicts (was RE: More random python observations from a perl programmer)
In-Reply-To: <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <001301beeacf$9819a560$712d2399@tim>

[tchrist]
>> GOTCHA: (medium)
>>     Slices in Python must be contiguous ranges of a sequence.
>>     In Perl, there's no such restriction.

[Michael Hudson]
> At some point Python will probably support `extended slice syntax'
> like so:
>
> range(0,10)[1:4:2] => [1,3]
>
> Numerical Python already does this.

[tchrist]
>> GOTCHA: (medium)
>>     You can't slice dictionaries at all in Python.  In Perl, it's
>>     easy to slice a hash, and just as sensible.

[Michael Hudson]
> Huh? What would that mean? There's no reason to suspect that the keys
> of a dictionary are ordered.

What Tom calls "slicing" here the NumPy folk would call gather/scatter
subscripting.  A list can be used as an index, e.g.

    thing[[7,2,3]] == [thing[7], thing[2], thing[3]]

Perl has something like that; Python does not.

can't-tell-the-subscripting-modes-without-a-scorecard-ly y'rs  - tim





From wtanksle at dolphin.openprojects.net  Fri Aug 27 18:09:58 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 27 Aug 1999 22:09:58 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> 	<7p49l5$uv7$1@nnrp1.deja.com> 	<37b634a2@cs.colorado.edu> 	<FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net> 	<37C552DF.9F90BA36@megsinet.net> 	<slrn7sblvo.p96.wtanksle@dolphin.openprojects.net> <14277.63776.733073.298192@dolphin.mojam.com> <37C60262.C6771B75@megsinet.net>
Message-ID: <slrn7se35m.sii.wtanksle@dolphin.openprojects.net>

On Thu, 26 Aug 1999 23:13:39 -0400, David Oppenheimer wrote:
>Where can I find info on Minotaur?

I don't remember, but try looking it up on http://freshmeat.net/.

>David O.

-- 
-William "Billy" Tanksley



From ketil at ii.uib.no  Wed Aug 25 05:16:00 1999
From: ketil at ii.uib.no (Ketil Z Malde)
Date: Wed, 25 Aug 1999 09:16:00 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <37c2a462.349531009@192.77.87.74> <935523275snz@vision25.demon.co.uk>
Message-ID: <KETIL-86u2pote5z.fsf@ketilboks.bgo.nera.no>

philh at vision25.demon.co.uk (Phil Hunt) writes:

> The Parrot tool has other potential advantages, e.g: 

> 1. you only have to learn one syntax, that can then be used for
> lots of languages and GUI libraries.

> 2. assuming the right import and export tools have been written,
> Parrot can allow you to use one GUI builder to output to a different
> GUI, for example you could design your GUI in Glade, and use Parrot
> to output it as a series of C++ files for KDE/Qt.

Or - you could extend Parrot to allow switching between GUIs on the
fly.  Parse the GUI spec, building the dialogs and other GUI
components, and voila! your application has a new "skin" (or "theme"
or whatever).

I suppose it might be easier to use an already interpreted language
like TCL to do this, create GUI as separate components that hook into
the application main code.  You'd probably want some business logic in 
the GUI scripts though, like disabling certain widgets depending on
the state in another, etc.

-kzm
-- 
If I haven't seen further, it is by standing in the footprints of giants



From fredrik at pythonware.com  Wed Aug 25 04:03:19 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 25 Aug 1999 08:03:19 GMT
Subject: xanadu and python
Message-ID: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com>

I doubt anyone have missed that Ted Nelson's Xanadu,
the original hypertext system, has finally been released
as open source:

    http://www.udanax.com/

what Python has to do with it?  well, go to the site and
find out for yourself ;-)

</F>




From bruce_dodson at bigfoot.com  Sat Aug 14 05:10:39 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Sat, 14 Aug 1999 06:10:39 -0300
Subject: OCX and PythonWin....
References: <7oq27u$th0$1@nnrp1.deja.com>
Message-ID: <%oat3.62670$jl.39620751@newscontent-01.sprint.ca>

On a similar subject.  I was experimenting with MapObjects (an ActiveX
control) and PythonWin.  It seems that every time I load it, PythonWin
crashes on exit (access violation or something).  I tried generating a quick
AppWizard test in VC++, and had no such problems there.  Also no problems in
VB, of course.

My application follows the Internet Explorer example in pywin/demos/ocx.
That example runs fine, but not with MapObjects.  Is there something that
needs to be done to tell pythonwin that this object should be torn down
after its window is closed?  An explicit DestroyWindow didn't help.  I might
try putting the control in a dialog resource and launching it into a
CFormView to do the doc/view thing and perhaps get better integrated with
MFC, but I'd really rather not.  Can anyone offer some words of wisdom?
(btw Mark if you're listening, there's a spot reserved on my bookshelf for
your upcoming Python/Win32 book.)

There is some benefit to this pain.  I am slowly coming to grips with the
way PythonWin wraps MFC.  A light turned on as I was writing this and
realized why the docs talk about PyCMDIChildWnd while the examples derive
from things like MDIChildWnd, why the latter isn't mentioned in the help
file whereas the former can't be found in the object browser.  But it's
6:00am.  After a good morning's sleep I will probably have forgotten what I
learned. :-)




From jb12345678 at my-deja.com  Sun Aug 22 11:35:09 1999
From: jb12345678 at my-deja.com (jb12345678 at my-deja.com)
Date: Sun, 22 Aug 1999 15:35:09 GMT
Subject: win32com: ambient properties
Message-ID: <7pp5bc$dnh$1@nnrp1.deja.com>

Hi,

I am experimenting with mshtml as a control (obtained from WebBrowser control
as HTMLDocument instance). I have a problem getting the mshtml control into
Editing mode. According to the mshtml specs I need to set an ambient property
in the container: AMBIENT_USER_MODE to false. The control should query the
container for this ambient property.

I don't know how to do this. It does not work when I set the property on the
HTMLDocument instance directly.

Does anyone have a hint?

-Jb


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From joe at strout.net  Mon Aug 30 14:31:36 1999
From: joe at strout.net (Joseph J. Strout)
Date: Mon, 30 Aug 1999 11:31:36 -0700
Subject: 'if foo' calls getattr?!?
In-Reply-To: <1276085477-52750363@hypernet.com>
References: <1276085477-52750363@hypernet.com>
Message-ID: <v04210108b3f07e26764e@[198.202.70.213]>

Hi Gordon,

>In which case "if foo is not None" will be the fastest and
>closest-to-your-intention test possible. (But "if foo != None" will
>get you right back to where you are now!).

Yep, that's what I'm looking for.  (I had tried "if foo != None" 
already, but I rarely use "is" and "is not", so it didn't even occur 
to me to try.)

Thanks,
-- Joe

,------------------------------------------------------------------.
|    Joseph J. Strout           Biocomputing -- The Salk Institute |
|    joe at strout.net             http://www.strout.net              |
`------------------------------------------------------------------'



From greg.ewing at compaq.com  Mon Aug 23 17:08:52 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 24 Aug 1999 09:08:52 +1200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <yaZu3.23$2y4.21931@news-dal.corridex.com>
Message-ID: <37C1B864.970C63DC@compaq.com>

Evan Simpson wrote:
> 
> A syntax for expressing explicit tuple
> unpacking has been proposed (fn(*t)), and it remains to be seen whether it
> will enter the language.

I'm working on an implementation of that, by the
way, so it may be possible for people to try it
out soon.

> There are no compound
> assignment operators since there are no assignment operators, period.

I don't think it's helpful to jump on that particular
terminological detail. The people who ask "why is there
no += operator" aren't using the word "operator" in its
precise technical Python sense.

> It's what you would expect if you assume floating point arithmetic at all
> times.  If you're operating on integers, it's quite wrong.  Explicitness
> wins, in Python.

I'd say quite the opposite - this is a rare case in
Python where IMplicitness loses! If the distinction
between integer and floating point division were
made explicit, by having different operators for
them, this would cause much less grief.

Greg



From tim_one at email.msn.com  Sat Aug 21 22:47:05 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 21 Aug 1999 22:47:05 -0400
Subject: More random python observations from a perl programmer
In-Reply-To: <7pmsdi$81u@ds2.acs.ucalgary.ca>
Message-ID: <000401beec48$9e4e88e0$cb2d2399@tim>

Skipping most of this, since I already posted my own long reply to Tom.

[Tom Christiansen, admits <wink> that Perl copies arrays]

[Neil Schemenauer]
> What happens when you have a large array and pass it to a
> function?  This seems pretty inefficient.  If the array is not
> copied then why are they so many rules about when something is
> copied and when it is not.

The array gets copied.  In Perl4 there were tricks you could play with
symbol tables to avoid the copy, or you could rely on dynamic scoping and
not pass anything; in Perl5 there's a general notion of "reference type",
and you can avoid the copy by passing a reference to the array.  The callee
needs to know it's a reference, though (indexing a (de)reference to an array
requires different syntax than indexing an array directly; a huge (to me)
advantage of Python's uniformity is that the caller can choose to pass the
original array or a copy as it pleases, and the callee doesn't know the
difference).

>From a Python person's point of view, Perl *always* copies, and in the case
of passing a reference simply copies a small indirection object.  Python
never copies, but that's hard to explain to a Perl person because "no copy
== construct a reference" to them, while that way of viewing it makes little
sense to a Python person.

there's-no-way-to-break-out-of-this<wink>ly y'rs  - tim





From arcege at shore.net  Mon Aug 16 20:44:33 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 17 Aug 1999 00:44:33 GMT
Subject: Redirecting Python Output
References: <rrh3pe9goqs55@corp.supernews.com>
Message-ID: <Rf2u3.315$mV.112571@news.shore.net>

Alberto Fonseca <alberto at totallygames.com> wrote:
: Hello, I am embedding Python into my app under Windows.  I would like to
: redirect Python's output so that I can just print it myself.  Has anyone
: done this?  What is the easiest way to do this?

Probably something like:

    import sys
    try:
      from cStringIO import StringIO
    except:
      from StringIO import StringIO
    sys.stdout = StringIO()
    sys.stderr = StringIO()

Each file output is seperate here, but could be the same to redirect
the output to both places.  This will allow you to retrieve it in your
embedded C program thusly:

    PyObject *sys_stdout, *str;
    sys_stdout = PySys_GetObject("stdout");
    str = PyObject_CallMethod(sys_stdout, "getvalue", NULL);
    fprintf(stderr, "Python output: %s\n", PyString_AsString(str));
    Py_DECREF(str);
    Py_DECREF(sys_stdout);

  -Arcege




From greg.ewing at compaq.com  Mon Aug 16 16:59:24 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 17 Aug 1999 08:59:24 +1200
Subject: Anyone else find Tk akward?
References: <37B34DBB.E5A79760@sprint.ca> <NeMs3.6611$gO1.175350@news2.giganews.com>
Message-ID: <37B87BAC.CE092017@compaq.com>

Christopher Browne wrote:
> 
> The Tk model of packing "objects" onto a "canvas" allows it to
> self-optimize, and lets the programmer not need to worry so much about
> those sorts of fiddly details.

This is good, except that if you just pack things
together in the simplest and most obvious way, the
results are awful -- everything is jammed together,
and nothing resizes the way you want. To get a layout
which looks good and works properly, you need to deal
with another set of fiddly details, namely the various
padding, filling and expanding options, which (for
me, anyway) seem to work in very unintuitive ways.

The idea of an automatic layout manager is as good
one, but I don't think that any of Tk's geometry
managers are a good implementation of it. I'm not
surprised that some people find them frustrating.

Greg



From hschroeder at gmx.net  Sat Aug  7 19:29:07 1999
From: hschroeder at gmx.net (Henning Schroeder)
Date: Sun, 8 Aug 1999 01:29:07 +0200
Subject: Tkinter -- tarred up documentation?
References: <934062911snz@vision25.demon.co.uk>
Message-ID: <3gfio7.ia4.ln@odin.henning-schroeder.de>

Phil Hunt <philh at vision25.demon.co.uk> wrote:

> On the Pythonware site (www.pythonware.com) there's a some Tkinter 
> documentation entitled "Introduction to Tkinter", and "Tkinter
> Class Reference pages". 
> Unfortunately these are stored as single HTML pages.
"Introduction to Tkinter" is also available as PDF on there website.
Nevertheless PIL documentation are single HTML pages :-(
Maybe someone has one of these download-utilities and could provide .tgz?

Henning
 



From thomsensven at my-deja.com  Fri Aug 13 04:29:07 1999
From: thomsensven at my-deja.com (thomsensven at my-deja.com)
Date: Fri, 13 Aug 1999 08:29:07 GMT
Subject: Problems importing Modules under win32
References: <7oud1c$vcd$1@nnrp1.deja.com>
Message-ID: <7p0l0d$i3j$1@nnrp1.deja.com>

Hi,

i found it, a __init__-File was missing.
Thanks to Lars Marius Garshol for the help...

Bye,

Sven Thomsen



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From chad at vision.arc.nasa.gov  Sun Aug 15 06:28:17 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Sun, 15 Aug 1999 10:28:17 +0000
Subject: Enough [Why fight when we can hoist a few at OSC?!]
References: <1277478426-31364719@hypernet.com> <37b58c4a@cs.colorado.edu>
Message-ID: <37B69641.35128019@vision.arc.nasa.gov>

Tom Christiansen wrote:

>    That man, along with Ian Clarke, have stirred up the biggest flame war
> in the Perl newsgroup that we're seen for a long time, inflicting their
> standard overrighteous prattle on us.

As one who put a flaming bag-o-turd on Tom's home doorstep (c.l.p.m,
that is), when I should not have contributed to such a bonfire of abuse
(although that "if 1:" trick can be useful :), I'd like to say that it is a
shame
this had to start so near to the O'Reilly conference.  I won't be able to make

it, but I hope others will not let this interfere with any potential
discussion
and socializing between the two "camps".

Anyone who buys Tom a beer for me (or an equivalent, depending on his
tastes), I'll pay them back.

Chad

 Uhhh, but don't everyone do it, or I'll go broke. :)





From tietz at psych.ucsb.edu  Wed Aug  4 14:52:07 1999
From: tietz at psych.ucsb.edu (Jerry Tietz)
Date: Wed, 04 Aug 1999 11:52:07 -0700
Subject: SGI C++ compile errors
Message-ID: <37A88BD7.3723671D@psych.ucsb.edu>

I am trying to compile an SGI C++ source
file which contains the call Py_Initializ() but I
get the following error messages:

"/usr/people/beall/dev/Python-1.2/Include/mymalloc.h", line 64:
error(3419):
          more than one instance of overloaded function "malloc" has "C"

          linkage
  extern ANY *malloc Py_PROTO((size_t));
              ^

"/usr/people/beall/dev/Python-1.2/Include/mymalloc.h", line 65:
error(3419):
          more than one instance of overloaded function "calloc" has "C"

          linkage
  extern ANY *calloc Py_PROTO((size_t, size_t));
              ^
Can anybody tell me how to fix this?

Thanks,
Jerry Tietz




From sjmachin at lexicon.net  Tue Aug 17 16:23:00 1999
From: sjmachin at lexicon.net (John Machin)
Date: Tue, 17 Aug 1999 20:23:00 GMT
Subject: Regarding the magical "or" operator
In-Reply-To: <000601bee6cb$a5e747e0$832d2399@tim>
References: <000601bee6cb$a5e747e0$832d2399@tim>
Message-ID: <19990817202744770.AAA404.73@max41104.izone.net.au>

A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 1300 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990817/2f0a09c1/attachment.bin>

From gary at onegoodidea.com  Tue Aug 24 08:40:17 1999
From: gary at onegoodidea.com (Gary O'Keefe)
Date: Tue, 24 Aug 1999 12:40:17 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu> <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net> <slrn7s2dtk.vsl.kaz@ashi.FootPrints.net> <37c21fae$0$216@nntp1.ba.best.com>
Message-ID: <37c28ac9.13320348@news.hydro.co.uk>

A keyboard was whacked upside Xah's head and out came:

[ a vast and rambling discourse snipped ]

>AppleScript only looks English, but does not have all the essential
>qualities of natural languages. Like Python and others, it's too consistent
>and ram some OOP or Object Models down your throat. Don't start me on Apple.
>When it comes to Apple, they always push for useless change. In the past
>years they degraded programers' social status with the oh--sooo-cute color
>iMacs and G3 towers, and in September they'll have this
>oh--sooo-cute-I'm-drooling iBook and AirPort technology probably semi-stolen
>from unix's RFC standards. (RFC is an abbreviation for Really Fucking
>Common)

IIRC, AirPort is licensed from Lucent. 

Anyway... damn straight! Back to the bad old days of gnawing out the
holes in punched cards with your canines, 'cos the hole punching
machine costs another $500,000, fits into a small warehouse, uses
enough electricity to electrocute England and weighs more than the
Earth itself. Programming that got respect. Real spit and bailing-wire
programming. None of this namby-pamby open-up-a-spreadsheet-record-
a-macro-and-knock-it-about-a-bit "programming" for us, eh?

How unfortunate that computers (and programming) have been brought to
the attention of the bungled and the botched by the Great Satan Apple
and their evil couterparts in Redmond. When the proletariat figure out
how easy all this stuff *really* is, then they're going to turn on us
and rip us to shreds like a pack of hungry wolves for charging them so
damned much all these years.

>As Larry has well said, Perl is a post-modern language. The language of the
>New Age era.

Pity all the hippies have got colostomy bags and Zimmer frames these
days. We're post-post-modern now, or hadn't you heard? (Turn up the
hearing aid... ;)

Gary

p.s. To the other poster (who I really cannot be bothered searching
for) who asked if the point of learning programming was just to earn
vast lumps of cold, hard cash at a F500 company: yes it is.
--
Gary O'Keefe
gary at onegoodidea.com

You know the score - my current employer has nothing to do with what I post



From cbbrowne at news.hex.net  Mon Aug 30 22:18:17 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Tue, 31 Aug 1999 02:18:17 GMT
Subject: Python Versus Whitespace (Re: Best language for graphical apps?)
References: <rs8v5gdrh9721@news.supernews.com> <h4hr9kqyzzl.fsf@libr-125.rcs.rpi.edu> <2vwvuijaxf.fsf@spls61.ccs.mt.nec.co.jp> <935778741snz@vision25.demon.co.uk> <7q98h5$n2815@george.sabre.com> <h4hiu5z2xoe.fsf@vcmr-19.rcs.rpi.edu>
Message-ID: <JXGy3.430$Jl.19931@news6.giganews.com>

On 28 Aug 1999 21:16:49 -0400, Collin W. Hitchcock
<hitchc at vcmr-19.rcs.rpi.edu> wrote: 
> On some earlier date, Christopher Browne <cbbrowne at hex.net> wrote...
>> There seem to be dogmatic positions about the evils of Python's
>> ability to use whitespace to signify control structures;
>> unfortunately, the dogmatic positions seem more often based on blind
>> faith than on tenable arguments.
>
>A list of tenable arguments why I hate white-space syntax dependent
>languages:
>
>   1) A whole slew of trouble with code that uses tabs.  Use your
>   imagination.  Read about the history of "make".

I've never had problems with makefiles, no, never...  :-)

>   2) For bracket grouped languages I prefer the indenting style:
>   
>         block_start
>         {
>           sub_statement
>           sub_statement   
>         }
>   
>   In Tcl it's a syntax error to put the opening bracket on a new
>   line.  In Python the brackets don't exist.  So now I'm looking at
>   someone else's hideous code with 240 column lines.  It's written so
>   that it's completely incomprehensible unless you can see whole
>   lines at a time, so you have to put your editor in wrap mode.  If
>   the brackets exist you can first make sure the code is indented
>   consistently and then look for opened and closed brackets in the
>   same column.  You can filter out all the line wrap garbage when
>   you're looking at block structure.

Analyzing Lisp code is Pretty Easy on a similar basis; the lack of
ambiguity is indeed attractive.

>   3) If you are trying to write a code generator, white-space syntax
>   dependent languages are a nightmare.  So much easier to produce the
>   code without worying about indenting and then run a second program
>   to indent it.

This is *not* one that I see as being a big problem.

In a brace-structured language, you need some structure that counts
braces. 

That parallels the need, in a Python generator, to have some structure
that counts the amount of whitespace in effect.

They're isomorphic to one another.

>4) You can introduce *hideous* bugs by cutting and pasting between
>two differently indented sections of code.  When white space
>doesn't matter, you just paste and tell your editor to reindent.
>
>None of these are deal breakers -- you can work around all of them.
>They just add to the general stress level.  I already have enough
>stress, thankyou.

As you say, these can be worked around.

Thank you for not merely making dogmatic comments; mere dogmatism is
more common in such discussions.
-- 
if (argc > 1 && strcmp(argv[1], "-advice") == 0) {
  printf("Don't Panic!\n");
  exit(42);
}
(Arnold Robbins in the LJ of February '95, describing RCS)
cbbrowne at ntlug.org- <http://www.ntlug.org/~cbbrowne/langscript.html>



From stalnaker at acm.org  Sun Aug 15 15:04:28 1999
From: stalnaker at acm.org (Max M. Stalnaker)
Date: Sun, 15 Aug 1999 12:04:28 -0700
Subject: Does Python run under Windows 2000? (build 2031)
References: <37B40D58.DD6AADC0@RobotVision.com>
Message-ID: <37B70F3C.1A194BB5@acm.org>

It works for me, but I am on build 2072, which I think of as RC1.  2031
sounds like beta 3.

In my configuration, win32all-125, in common with some other software, puts
some dlls into the c:\documents and settings subdirectories, unexpectedly.
This results in a need to use regedit and then manually register some com
components.  Not all users have reported this to be true.

I spent $50 for registration of win32all with Mark Hammond and he provided
excellent technical support for me in getting past the regedit and com
registration difficulties.  This is very nice to have on an unsupported os.

python itself showed no difficulties in installing.





"James E. Green" wrote:

> I'm running Windows 2000 RC1 (build 2031, the very latest).  Will Python
> run under this OS?  Only NT4 is mentioned on the Python web page.
>
> Thanks, Jim Green

--
Max M. Stalnaker mailto:stalnaker at acm.org http://www.astarcc.com
253-565-2366 fax 253-565-0892





From e_l_green at hotmail.com  Sun Aug  1 22:52:43 1999
From: e_l_green at hotmail.com (Eric Lee Green)
Date: Sun, 01 Aug 1999 19:52:43 -0700
Subject: RMI and Python
References: <Pine.GSO.4.10.9908011131490.25758-100000@xmission.xmission.com>
Message-ID: <98A6BBA3E6835B16.321CAF48B0B697EF.8965F6D61944B70C@lp.airnews.net>

Tim Egbert wrote: 
> Java is able to do remote method invocation (RMI) which is a neat concept.
> It is the basis for Jini, another neat idea.  Does Python have anything
> like RMI?

It's easy enough to create one, using the SimpleTCPServer class. In
fact, I have done it. As soon as I get a chance to re-visit the code and
clean up some boojums I will toss it onto my web site (it currently is
functional and working, but there are some known denial-of-service
problems that I wish to tackle before considering it halfway finished). 

If you're wanting to do it youself, think HTTP -- HTTP is at its heart a
remote method invocation mechanism, you have the HTTP libraries
available, you have the BaseHTTPServer available so that you can add
your own methods (HTTP is NOT restricted to GET and POST, though that's
all that most HTTP servers understand!), and Python's 'pickles' make it
very easy to wrap up data to pass it to a remote server. 

-- 
Eric Lee Green    http://members.tripod.com/e_l_green
  mail: e_l_green at hotmail.com
                    ^^^^^^^    Burdening Microsoft with SPAM!



From tchrist at mox.perl.com  Sat Aug 14 17:52:34 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 14 Aug 1999 14:52:34 -0700
Subject: Regarding the magical "or" operator
Message-ID: <37b5d712@cs.colorado.edu>

I was wondering how often most python programmers make use of that
remarkably nifty property of the logical operators, in which they
produce not merely 0 or 1, but the value is last looked at when it
decided whether the expression was true or false.

For example:

	print [] or "Some"
    Some

	print "Peter" and "Paul"
    Paul

	print (20 - 4*5) or None
    None

My question is this: have you ever had an urge for it (or a variant)
to care about the difference between None and any other false value?
I am certainly *not* asking for such a thing.  I'd just like to know
whether anyone ever cared.  For example, given these three functions:

	def f():
	    return []

	def g():
	    return None

	def h():
	    return ["Some"]

Consider these prints:

	print f() or h()
    ['Some']

	print g() or h()
    ['Some']

The idea would be a version of "or" that would return the first non-None
value.  For example, supposing this were called "nor" (silly name,
but this is *not* a proposal):

	print f() nor h()
    []

	print g() nor h()
    ['Some']

The argument is that this would be used not with prints, but with
assignments, as in

    a = f() nor g()

The reason I ask is not because I actually want such a thing.  Gosh, no.
I'm rather uncomfortable with the idea.  I'm just trying to gather some
understanding and perspective for use with the folks who *would* like
such an oddity. :-)

--tom
-- 
Mister Catbert, the company is trying to force me to use a different kind
of computer.  You're the human resources directory. what are you doing
to sop this religious persecution?!  What every happened to "Diversity"??
The longer you verk here, diverse it gets. next.  --Scott Adams, "Dilbert"



From emile at fenx.com  Sun Aug  1 13:28:07 1999
From: emile at fenx.com (Emile van Sebille)
Date: Sun, 1 Aug 1999 10:28:07 -0700
Subject: spawnv
References: <19990801171952.JFVG5731@default>
Message-ID: <1w%o3.4414$VL2.474450@news.direcpc.com>

Hi Craig,

This works on WIN95:

os.system(r"start /w c:\windows\notepad.exe")

It leaves a dos window displayed however.  If you change the properties
of the dos window to minimized, the next time through it won't display.
I imagine it's possible to preset this, but I haven't tried.  Does
anyone know how to stop this or get around it?

--

Emile van Sebille
emile at fenx.com
-------------------


Craig Curtis <c.curtis at worldnet.att.net> wrote in message
news:19990801171952.JFVG5731 at default...
> I am trying to spawn c:\windows\notepad.exe from a Python program.  I
want
> the python script to wait for Notepad to exit.
> How can I do this?
>
> I have tried all combinations I can think of using execv, spawn,
system and
> cannot get it to work.
>
> Could someone please post the exact command sequence to type in.
>
> Thanks in advance,
> Craig Curtis
>
>





From goertzen at norscanNOSPAM.com  Thu Aug 26 17:58:35 1999
From: goertzen at norscanNOSPAM.com (Daniel Goertzen)
Date: Thu, 26 Aug 1999 21:58:35 GMT
Subject: static python build
Message-ID: <fMix3.1715$1i4.132844@typhoon.mbnet.mb.ca>

How do I build python statically on linux so I don't need any of the shared
libraries?

Dan.





From fdrake at acm.org  Wed Aug  4 13:45:56 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Wed, 4 Aug 1999 17:45:56 GMT
Subject: building embedded classes as in C/C++
In-Reply-To: <37A86B09.CB163857@psych.ucsb.edu>
References: <37A78579.9856D16@psych.ucsb.edu>
	<al8907sqh81.fsf@sirppi.helsinki.fi>
	<37A86B09.CB163857@psych.ucsb.edu>
Message-ID: <14248.31828.719428.809970@weyr.cnri.reston.va.us>

Andy Beall writes:
 > Here, 'p' does not seem to be class-wide while 'pos' and 'ang' are.  Why
 > the discrepancy?

Andy,
  The 'p' set in the class statement is class-wide.  Setting 'p' on an 
instance creates a new instance-specific binding for 'p', which is
retrieved on subsequent attribute accesses since it overrides the
class-wide value.  If you accessed it as Object.p, the original value
would still be available.
  This is a situation where the "binding" and "namespaces" terminology
is pretty important and makes it easier to understand.  There really
aren't variables in the C/C++/lots-of-others sense. Sections 9.1 and
9.2 in the tutorial may be worth reviewing here:

	http://www.python.org/doc/current/tut/node11.html


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From wtanksle at dolphin.openprojects.net  Mon Aug 23 22:18:53 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 24 Aug 1999 02:18:53 GMT
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan>
Message-ID: <slrn7s408d.gvn.wtanksle@dolphin.openprojects.net>

On 23 Aug 1999 23:32:44 +0200, forcer wrote:
>Hi there.

>Lately, i've been thinking about Python. While it is an
>exceptionally clean language, it has one very obvious wart:
>The ``pass'' statement.

I disagree that 'pass' is an obvious wart.  I think it's very nice.

>To quote the reference manual:

>   pass is a null operation -- when it is executed, nothing
>   happens. It is useful as a placeholder when a statement is
>   required syntactically, but no code needs to be executed, for
>   example:

>   def f(arg): pass    # a function that does nothing (yet)

>one could write

>	if foo:
>	bar

>Syntactically this does not introduce any ambiguity. The only
>problem i see is that the "pass" version is more readable.

Precisely.  You're suggesting a change with the sole purpose of reducing
redundancy.  The problem is that humans _need_ redundancy.  Not only does
redundancy help us to read the code, it also helps the computer to find
our errors.

Another example of redundancy in Python -- and one with just as much
chance of being changed -- is the ':' symbol.  Take a look at the following:

if foo: pass
bar

if foo pass
bar

if foo
  pass
bar

As you see, the colons aren't necessary.  They're only there as a
convenience to the programmers and readers.

>       --forcer

-- 
-William "Billy" Tanksley



From .xanafonte at mail.pt  Tue Aug 31 19:59:54 1999
From: .xanafonte at mail.pt (Xana)
Date: Tue, 31 Aug 1999 23:59:54 GMT
Subject: position on the screen
Message-ID: <37cc6b98.873539@news.telepac.pt>

I'm working with PythonMegaWidgets and i have a problem using
Pwm.dialog boxes.
Can anyone tell me if  (and how) I can define the position in wich the
window will appear in the screen ?

thanks in advance


Xana



From python-list at teleo.net  Sat Aug 21 03:21:12 1999
From: python-list at teleo.net (Patrick Phalen)
Date: Sat, 21 Aug 1999 00:21:12 -0700
Subject: More random python observations from a perl programmer
References: <7ple6r$kf0@ds2.acs.ucalgary.ca>
Message-ID: <9908210028270C.00619@quadra.teleo.net>

On Fri, 20 Aug 1999, Neil Schemenauer wrote:
> As no one seems to have written a reply to all of Tom's points, I
> will make an effort.  Some of this information has been already
> presented in other posts.  Hopefully this will be an accurate
> summary.

[deliciously well-balanced summary snipped]

Accurate indeed; and a real pleasure to read. This is the Usenet at
its best.

You have such a talent for clear explanation that I would hope
you might try your hand at writing some succinct Python tutorials for
beginners. I'm sure they would earn a place beside Guido's own
Tutorial, Learning Python, and the Grimoire.




From rodzilla2000 at my-deja.com  Wed Aug  4 23:22:31 1999
From: rodzilla2000 at my-deja.com (rodzilla2000 at my-deja.com)
Date: Thu, 05 Aug 1999 03:22:31 GMT
Subject: Stupid question about .EXE
References: <7o93uk$sjrn$1@reader2.wxs.nl>
Message-ID: <7ob01f$ghn$1@nnrp1.deja.com>

In article <7o93uk$sjrn$1 at reader2.wxs.nl>,
  "Rik" <rik-007 at dds.nl> wrote:
> Hi everybody,
>
> I've got a (stupid) question. I'm learning myself Python now, but I
still
> have a question:
> How can you make .EXE files with python??? I'm using Python 1.5.2
(final)
> for Windows.
>
> Greets
>
> Rik
>

Hi, I'm new to python too...

I'm sure that python itself cannot make .EXE files, but I have read
somewhere that you can encapsulate a python script into an EXE...

Don't know what the name of the utility is, or if it's executable
itself or is c source code to allow the making of a python .EXE

Does anyboy know what the name of the utility is and where it's located?
I read somewhere about it, but can't remember....


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcane*SPAM*dragon at home.com.invalid  Thu Aug 19 10:10:50 1999
From: arcane*SPAM*dragon at home.com.invalid (Chris Herborth)
Date: Thu, 19 Aug 1999 14:10:50 GMT
Subject: CVS classes in Python
References: <37B9969B.BB40C25F@primenet.com> <934914414snz@vision25.demon.co.uk>
Message-ID: <KfUu3.10629$H6.1959198@news1.rdc1.on.wave.home.com>

According to Phil Hunt <philh at vision25.demon.co.uk>:
> In article <37B9969B.BB40C25F at primenet.com>
>            yerny at primenet.com "Bryan Jernigan" writes:
> > I'm a newbie to Python and I would like to create a Python script to
> > maintain a distributed CVS CM system.  Before I re-invent the wheel, I
> > was wondering if anyone has some decent CVS classes and/or functions.
> > If so, please drop me a line.
> 
> Why not have your Python classes acting as a front end to CVS or PVS?

This is what I did for some RCS-aware SGML parsers I've written in Python;
they just call out to the RCS commands when necessary, via os.system().
Not very sophisticated, but it works. :-)

-- 
----------================================================----------  _
Chris Herborth, R&D, Technical Writing                               | \  _
Arcane Dragon Software                 Arcane Dragon -==(UDIC)==-    | < /_\
arcane dragon at home dot com          DNRC Holder of Past Knowledge |_/ \_



From andrew at starmedia.net  Thu Aug 19 18:31:43 1999
From: andrew at starmedia.net (Andrew Csillag)
Date: Thu, 19 Aug 1999 22:31:43 GMT
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com>
		<37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu>
Message-ID: <37BC85CF.7AA4A23B@starmedia.net>

Tom Christiansen wrote:
> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.python,
>     drew_csillag at geocities.com writes:
> :There are the python manuals which are excellent, available at:
> :http://www.python.org/doc/ (linked to from the homepage)
> :available in your choice of HTML, PostSript, PDF, Info, LaTeX and
> :windows help.
> 
> The fact that they did icky windows help and not man pages says
> something significant.

Yes, that python has a decent sized population of windows users.  I'm no
windows guy but PythonWin is way cool from what I've seen (IDE, cool COM
support, etc.).  

-- 
"Programmers are just machines that convert coffee to executable code"
   -Unknown

"Drew Csillag" <drew_csillag at geocities.com>



From shecter at darmstadt.gmd.de  Thu Aug 19 16:13:21 1999
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Thu, 19 Aug 1999 22:13:21 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <yaZu3.23$2y4.21931@news-dal.corridex.com>
Message-ID: <37BC6561.6B4CC7@darmstadt.gmd.de>

Evan Simpson wrote:
> ...Python programmers appear to prefer hyperlinked over man-style
> documentation...

Well, I don't think we should base assumptions on the posts of one
vocal advocate, but - 

I'd say it has to do with the language itself:

Perl is "old school", along with awk, and sed:  The complexity of the
system is in the language definition itself.  Pure OO/Component-like
libraries are a secondary source for features.  In this environment,
having docs that read linearly, more like a book, may be the best
solution.

Python is "new school", along with Java and Smalltalk (years ahead of
its time).  The language is intended to be simple, and the complexity
is in the libraries.  In this case, having highly structured,
hyperlinked refs may be the best way to go.  

- Robb



From gmcm at hypernet.com  Mon Aug  9 23:46:07 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 9 Aug 1999 22:46:07 -0500
Subject: Redirect stderr
In-Reply-To: <87lnbk8kow.fsf@linuxcare.com>
Message-ID: <1277869823-7876266@hypernet.com>

David N. Welton wrote:

> I wish to do this:
> 
> #!/usr/bin/python
> 
> import sys
> import os
> 
> f = open("goober", "w")
> 
> sys.stderr = f
> sys.__stderr__ = f
> 
> os.execv("/bin/ls", ["/bin/ls", "sdfsdfdsfds"])
> 
> and have it dump the errors into 'goober'.  I'm having some
> difficulties figuring out exactly why this isn't working, or how I
> ought to be doing this.

Because sys.stderr is a Python file object, which (by default) uses 
C's stderr. Reassigning it doesn't change C's assignment. If you want 
to twiddle file descriptor 2, you can do it with os.close(2) etc. Or 
be lazy and just redirect it on the command line to the subprocess.

- Gordon



From ngps at madcap.dyn.ml.org  Fri Aug 20 11:15:34 1999
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 20 Aug 99 15:15:34 GMT
Subject: ANNOUNCE: M2Crypto 0.01
Message-ID: <37bd7116.0@news.cyberway.com.sg>

Hello,

I am pleased to announce the release of M2Crypto 0.01;
the package is available at http://www.post1.com/home/ngps

Following is the README:

 The M2 Python Crypto Toolkit 0.01
-----------------------------------

The M2 Python Crypto Toolkit (M2Crypto) ==
	Python		www.python.org
+	OpenSSL		www.openssl.org
+	SWIG		www.swig.org

Python is an interpreted, dynamic-typing, object-oriented programming 
language that is often compared to Perl, Scheme, Tcl or Java. SWIG -
Simplified Wrapper and Interface Generator - provides glue to link C/C++
libraries with the above-mentioned languages.

M2Crypto presents a Python interface to OpenSSL, via SWIG.

M2 stands for "me, too!"*

This release includes the following: DH, RSA, DSA, RC4, MD5, SHA1,
RIPEMD160, and the EVP interfaces for message digests, HMACs, and 
symmetric ciphers. It requires Python 1.5.2, OpenSSL 0.9.4, and,
optionally, SWIG 1.1p5 or later. It has been tested under FreeBSD 
2.x, Redhat Linux 5.2, and WinNT4. It should run anywhere Python 
and OpenSSL run.

M2Crypto is released under a license similar to Python's. See LICENSE
for details.


For now, M2Crypto is intended to be a prototyping tool; it is _not_ 
for writing production crypto software.

In particular, note the following caveats:

1. There will be memory leaks, because of pointer ownership contention
between Python and SWIG. (These would be due to my misprogramming.)

2. There is no memory locking/clearing for keys, passphrases, etc.

3. The PRNG has no interface for CS**; it is unlikely to be CS***.

4. AFAIK, Python and OpenSSL have not been subjected to the full attention 
of the Bugtraq crowd. M2Crypto's handling of active hostile input is 
probably suspect.

Of course, I hope to address these in future.

Meanwhile, have fun! Your feedback is welcome.


Ng Pheng Siong
ngps at post1.com


* Similar software are Marc-Andre Lemburg's mxCrypto, and two (?)
Python interfaces to the SSL portion of SSLeay/OpenSSL.

** Continuous seeding.

*** Cryptographically strong.

-- 
Ng Pheng Siong <ngps at post1.com>




From mwh21 at cam.ac.uk  Thu Aug 19 13:06:03 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Aug 1999 18:06:03 +0100
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk>

I'm going to be quite picky in some of what follows.

Tom Christiansen <tchrist at mox.perl.com> writes:

> COOLNESS:
>     Python's print() function is more like the Perl debugger's
>     "x" command; that is, it's recursive and pretty-printed.

Python doesn't have a print() function. It has a `print' statement. 
 
> GOTCHA: (high)
>     Local variables are never declared.

*Nothing* is ever delcared in Python.

> GOTCHA: (medium)
>     If you have a function that expects two arguments, such as:
> 	def fn(x,y):
> 	    return x + y
>     and you have a tuple of the two elements:
> 	t = (1,2)
>     You can't just call that function:
> 	print fn(t)
>     Or you get a "TypeError: not enough arguments; expected 2, got 1"
>     error.  You have to use this apply() function to get around this
>     issue:
> 	print apply(fn,t)
> 	3

Well, no. How would you like to pass a tuple to a function?
 
> GOTCHA: (medium)
>     Perl's hex() function is the opposite of Python's.  Perl's hex
>     function converts 22 into 37.  Python's hex function converts 37 into
>     0x22.  Oh my!  Like hex, Python has oct() completely backwards from
>     Perl.  Python's oct(44) returns 054, but Perl's oct(44) return 36.

This would have to do with strings not being automatically coerced
into numbers, methinks.
 
> SIMILARITY:
>     Constructors and destructors (__init__ and __del__) aren't 
>     automagically called in parent classes in Python.  Neither
>     in Perl.

This sucks, IMHO.

> GOTCHA: (medium)
>     Things that return lists can't be used where a tuple is expected.
>     A function that returns a list must be coerced into a tuple to 
>     use this, though.
> 	def fn: return [1,2]
> 	print "this %d is %d here\n" % fn()  # ILLEGAL
> 	print "this %d is %d here\n" % tuple(fn())
>     The illegal part points out that this is an 
> 	TypeError: illegal argument type for built-in operation
>     Which isn't very helpful.  

I think this may have changed in the CVS sources. <quick check>, Oh
no, it hasn't. It should.
 
> GOTCHA: (high)
>     Python has no manpages!  The horror!!!!!!!!!!!!!!!!!!!!!!
>     ENODOC

There's info format if you look for it and you have to have a CLI.

> GOTCHA: (low)
>     Often Python's error messages leave something to be desired.
>     I don't know whether 

Is this a joke? This is improving.

> GOTCHA: (medium)
>     All ranges are up to but *not including* that point.  So range(3)
>     is the list [0,1,2].  This is also true in slices, which is the
>     real gotcha part.  A slide t[2:5] does not include t[5] in it.

This does mean that range(a,b) or t[a:b] has b-a elements in it.

> DISSIMILARITY:
>     This also means that instead of substr() in Perl, you slice a string
>     as in 
> 	s = "string"
> 	print s[2:4]
>       ri
>     Yes, that's all you got.  Strange, eh?  See below on ranges.

What were you expecting?

> GOTCHA: (medium)
>     Slices in Python must be contiguous ranges of a sequence.
>     In Perl, there's no such restriction.  

At some point Python will probably support `extended slice syntax'
like so:

range(0,10)[1:4:2] => [1,3]

Numerical Python already does this.
 
> GOTCHA: (medium)
>     You can't slice dictionaries at all in Python.  In Perl, it's
>     easy to slice a hash, and just as sensible.

Huh? What would that mean? There's no reason to suspect that the keys
of a dictionary are ordered.

> GOTCHA: (high)
>     As we saw with lists, because everything is a reference, and there's
>     no way to dereference that reference, this means that again there
>     is also no built-in, intuitive way to copy a dictionary.  Instead, 
>     the suggested work-around is to write a loop:
> 	new = {}
> 	for key in old.keys:
> 	    new[key] = old[key]
>     But this is guaranteed slower, because it's not at the C level.

new = old.copy()
 
> GOTCHA: (medium)
>     There's no way to set up a permitted exports list.  The caller may have
>     anything they ask for.  

This is a general Python convention; you *can* shoot yourself in the
foot if you try hard enough. Means the debugger can be written in
Python, for one thing.
 
> COOLNESS:
>     DBM files seem (?) to automatically know about nested datatypes.   

Ah, that's pickle. pickle is cool.

> GOTCHA: (high)
>     Scopes don't nest in Python, but they they do in Pascal, Perl, or C.
>     This is supposedly "simpler", but it's very suprising in many ways:
> 	x = 1			# global
> 	def fn1(): 
> 	    x = 10		# implicit local
> 	    def fn2():
> 		print x		# whose?
> 	    fn2()
> 	fn1()
>     The answer is that it prints 1, because fn2 has no x in its local
>     scope, so gets the global.  The suggested work-around is
> 	    def fn2(x=x):
>     using default parameters.

Yes this is a bit annoying. The Right Way to attach state to a
function is a class, of course. Not that that helps much.
 
> GOTCHA: (low)
>     List objects have built-in methods, like 
> 	l.append(x)
>     But string objects don't.  You have to import
>     from the string module to get them, and then they're
>     functions, not methods.

This will change in 1.6.
 
> GOTCHA: (low)
>     You have insert and append methods for lists, but only
>     a del function.  

del statement.

> GOTCHA: (high)
>     Because you can't use readline() to get a number, people seem to enjoy
>     calling eval() just to get a string turned into a number:
> 	import sys
> 	str = sys.stdin.readline()
> 	num = eval(x)
>     This is scary.  

Do people really do that? I don't. It can be made safe:

num = eval(string,{'__builtins__':{}})

but that's a bit hairy. Use float instead.

>     Even scarier is the propensity for calling input(),
>     which auto-eval()s its input to make sure it's the right "type".
>     (Funny that a soi-disant "typeless language" should be so danged
>     picky about this.)  

Python is more "dynamically typed" than "typeless", I'd say.

> GOTCHA: (low)
>     Regexes default to emacs style, not egrep style!

Not any more. import re.

> GOTCHA: (low)
>     Python's eval() only works on expressions, not full code blocks full
>     of statements.  You need exec() for the whole thing.

Writing exec() suggests it's a function; it isn't, it's a statement.

> GOTCHA: (medium)
>     Anything that python doesn't like, it raises an exception about.
>     There is no fail-soft.  Even non-exception issues raise exceptions.
>     It's pervasive.  K&P curse languages that force people who want
>     to open() a file to wrap everything in exception code, saying that
>     "failing to open a file is hardly exceptional".  

This is a bit of a flaw, IMHO. Mind you, it forces you to pay
attention.

> GOTCHA: (medium)
>     You can't just interchange tuples and lists in python the way
>     you can lists and arrays in Perl.  This is an error:
> 	import sys  # otherwise, no argv for you, buddy
> 	print "First arg %s and second %s" % sys.argv[1:3]
>     because you need
> 	print "First arg %s and second %s" % tuple(sys.argv[1:3])

This has been changing over the last year or so. I expect the trend
will continue.

> COOLNESS:
>     I believe that Python checks the prototype signature on methods
>     as well as on functions.  I wonder whether Tim Bunce's ideas run-time
>     evaluation of prototypes for Perl might be able to do this.

You mean perl doesn't? Ouch.

> QUESTION:
>     What is and what is not thread safe?

Everything is, but in a crude way. There's a global interpreter
lock. This isn't the best of situations.

I hope this post has raised the signal-to-noise ratio in the
perl/Python debate (it could hardly lower it).

Michael



From I.Clarke at strs.co.uk  Mon Aug 16 07:41:57 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Mon, 16 Aug 1999 12:41:57 +0100
Subject: Enough [Was Re: Why use Perl when we've got Python?!]
References: <1277478426-31364719@hypernet.com> <37b58c4a@cs.colorado.edu>
Message-ID: <37B7F905.B7FF8DE0@strs.co.uk>

> That man, along with Ian Clarke, have stirred up the biggest flame war
> in the Perl newsgroup that we're seen for a long time, inflicting their
> standard overrighteous prattle on us. 

That is rich coming from the guy that sets up automatic monthly emails
to people because he doesn't like their quoting style.  As for stirring
up the flame war, Tom has posted around 60 messages on that thread, more
than most if not all others, many of them rude, patronizing, and
inflammatory.  I have posted around 20, and in all cases I remain polite
and respectful to the people I am posting to (even, in most cases, Tom).

If anyone would like some samples of Tom's postings I would be happy to
supply (he sent 'courtesy' carbon copies to my email address of most of
his utterances).

To describe this as hypocrisy is an understatement.

Ian.



From MHammond at skippinet.com.au  Thu Aug 12 09:45:11 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 12 Aug 1999 23:45:11 +1000
Subject: Problems importing Modules under win32
References: <7oud1c$vcd$1@nnrp1.deja.com>
Message-ID: <7ouj5c$jja$1@m2.c2.telstra-mm.net.au>

Maybe an early version of sax with a later version of Python?

Is there an __init__.py in the xml and sax subdirectories?

What do:
>>> import xml
>>> import xml.sax

Yield?

Mark.


thomsensven at my-deja.com wrote in message <7oud1c$vcd$1 at nnrp1.deja.com>...
>Hi,
>
>I wanted to try out saxlib, but I can't import it under Windows NT:
>
>dir-structure:






From n89553 at squid.uni-paderborn.de  Fri Aug  6 10:15:03 1999
From: n89553 at squid.uni-paderborn.de (Mirko Liss)
Date: Fri, 06 Aug 1999 16:15:03 +0200
Subject: Stupid Question?
References: <Bioq3.214$6E5.3247@typ12.nn.bcandid.com>
Message-ID: <37AAEDE7.86687CF4@squid.uni-paderborn.de>

jim moore wrote:
> 
> Is there a bulit in function that will give you N evenly spaced numbers
> between X1 and X2.  For example: 10 numbers between 0 and 1 ([0.1, 0.2, 0.3,
> ...])
> It would be similar to MATLAB's linspace.
> How about a function similar to range but allowing a floating point step?
> Seems like there would be one out there, so I thought I'd check before I
> wrote one.

If you want to do numerics in python, you might want to install
the LLNL-Distribution. It provides matrix-oriented math like Matlab and
there's
good documentation.

regards,

Mirko

-- 
M Li?, <n89553 at hrz.uni-paderborn.de>



From fredrik at pythonware.com  Thu Aug 26 02:38:38 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 26 Aug 1999 06:38:38 GMT
Subject: problem with im.show()
References: <7q1f58$r7v$1@tribune.mayo.edu>
Message-ID: <007301beef8d$a221bda0$f29b12c2@secret.pythonware.com>

Tom Kinter <tmk at mayo.edu> wrote:
> >>> im.show()
> TypeError: illegal argument type for built-in operation

it's a bug.  here's a fix:
http://www.python.org/pipermail/image-sig/1999-August/000826.html

</F>




From phd at emerald.netskate.ru  Mon Aug  9 04:27:08 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Mon, 9 Aug 1999 12:27:08 +0400 (MSD)
Subject: International characters and Python's string functions on Linux
In-Reply-To: <MPG.1215b842a85c7d2c989682@news.mop.no>
Message-ID: <Pine.LNX.4.04.9908091226010.30922-100000@emerald.netskate.ru>

On Sat, 7 Aug 1999, Alexander Staubo wrote:
> > import locale
> > locale.setlocale(locale.LC_ALL, "")
> > 
> Good idea, but:
> 
> Python 1.5.2 (#9, Jul  2 1999, 04:33:02)  [GCC 2.8.1] on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import locale
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "/usr/local/lib/python1.5/locale.py", line 4, in ?
>     from _locale import *
> ImportError: No module named _locale
> >>> 
> 
> This is Slackware Linux 2.2/libc6. Should I configure Python with some 
> kind of magic parameter for the locale stuff to work?

   Just recompile python and enable _locale module (or compile _locale
yourself).

> -- 
> Alexander Staubo             http://www.mop.no/~alex/
> "Elvis skipped around the car and threw open the trunk.
> 'Hoopla!' he was heard to say, 'This is a 7.62mm M134 General
> Electric Minigun. Up to 6,000 rounds per minute. 7.62mm X 51 shells.
> 1.36kg recoil adaptors. Six muzzle velocity of 869m/s.'
> 'Oh,' said Rex, 'So that's what it is.'"
> --Robert Rankin, _They Came And Ate Us: Armageddon 2: The B-Movie_

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From greg.ewing at compaq.com  Wed Aug 18 19:09:01 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 19 Aug 1999 11:09:01 +1200
Subject: representation of unbound methods - strange behaviour
References: <37bbecf1.100354231@news.omnilink.de> <14266.59462.274125.746880@dolphin.mojam.com>
Message-ID: <37BB3D0D.A66BAE0E@compaq.com>

Skip Montanaro wrote:
> 
> I believe it is constructing the method name on-the-fly using the names it
> finds in the class and function.

Yep. A slightly different test reveals
what is going on:

def h():
    pass

class A:
    pass

A.f = h
A.g = h
print A.f
print A.g

Output:

<unbound method A.h>
<unbound method A.h>

Greg




  It's not using the attribute name.
> 
>     >>> class A: pass
>     ...
>     >>> def foo(): pass
>     ...
>     >>> A.f = foo
>     >>> A.g = foo
>     >>> A.f
>     <unbound method A.foo>
>     >>> A.g
>     <unbound method A.foo>
> 
> Skip Montanaro  | http://www.mojam.com/
> skip at mojam.com  | http://www.musi-cal.com/~skip/
> 847-971-7098



From rav at smo3k.z.pl  Wed Aug  4 10:05:36 1999
From: rav at smo3k.z.pl (Rafal Smotrzyk)
Date: Wed, 04 Aug 1999 14:05:36 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk>
Message-ID: <QMXp3.5809$S3.82420@news.tpnet.pl>

#Simple but working

import re

class Response:
   def __init__(self):
      self.HTMLDocument=''
   def write(self,adata):
      self.HTMLDocument=self.HTMLDocument+adata

def GetTextAsHTMLText(s,repldict={}):
   for akey in repldict.keys():
      c=re.compile('\<\%\= *'+akey+' *\%\>', re.I)
      s=c.sub(repldict[akey],s)
   c=re.compile('\<\%', re.I)
   s=c.sub('""")',s)
   c=re.compile('\%\>', re.I)
   s=c.sub('Response.write("""',s)
   s='Response.write("""'+s+'""")'
   aresponse=Response()
   gdict={'Response':aresponse}
   exec s in gdict,repldict
   return aresponse.HTMLDocument

def DoTest():
   ahtmlsource=r"""
<HTML>
<BODY>
<%
Response.write('<h1>'+ScriptName+'</h1>\n')
Response.write('<table>')
x=0
while x<10:
   s='<tr><td>%s</td><td>%s</td></tr>\n'%('d1_'+`x`,'d2_'+`x`)
   Response.write(s)
   x=x+1
Response.write('</table>')
%>
<p>
<%
for i in range(10):
   Response.write(`i`+',')
Response.write('<br>')
%>
Test line 1 = <%= Test1 %>
Test line 2 = <%=Test2%>
</p>
</BODY></HTML>
"""
   adict={'Test1':'this is page constant','Test2':'another
constant','ScriptName':'This is script name'}
   s=GetTextAsHTMLText(ahtmlsource,adict)
   print s

if __name__=='__main__':
   DoTest()

# Rafal Smotrzyk






From che at debian.org  Tue Aug 24 16:26:37 1999
From: che at debian.org (Ben Gertzfield)
Date: 24 Aug 1999 13:26:37 -0700
Subject: Python equivilent of pack()?
References: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com> <14274.61530.251401.762453@weyr.cnri.reston.va.us>
Message-ID: <ytt6725sz5u.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Fred" == Fred L Drake, <fdrake at acm.org> writes:

    Bit> My question is this: is there a functional equivilent to the
    Bit> Perl pack() and unpack() functions in Python? Or even better,
    Bit> is there an easier way to do what I want with Python? If not,
    Bit> how many people think this would be

    Fred>   Take a look at the struct module; the pack() and unpack()
    Fred> functions are exactly what you're looking for!  (What names
    Fred> were you looking up in the index of the Library Reference?
    Fred> I'll add them to the index if they make sense to others.)

Fred, I think the problem is that the link to the index of all
functions in the library reference is hidden way down at the bottom of
the page, and as a very non-descript button hidden at the upper-right
hand corner of the page. In fact, I had no idea until just now that
such an index even existed for Python!

I think it must not be too uncommon for folks to miss out on the fact
that there's a function index to the library reference. Perhaps there
should be a big fat bold link to it at both the top and bottom of the
page.

It would also be a good idea to use better icons than 'toc' 'm' and
'i' at the top of the page; I know I completely overlooked them until
just now because they weren't very noticeable.

Ben

-- 
Brought to you by the letters R and Z and the number 4.
"If you turn both processors off, you will have to reboot."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From tismer at appliedbiometrics.com  Thu Aug 12 07:59:24 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 12 Aug 1999 11:59:24 GMT
Subject: Macro preprocessor.
References: <slrn7r291c.137.kaz@ashi.FootPrints.net>
Message-ID: <37B2B71C.E380A71@appliedbiometrics.com>


Kaz Kylheku wrote:
[description of a cute macro processor skipped]

> Okay, enough. :) I'll probably get hate mail now about being off topic.

No hate mail, but some thoughts:

I can think of macro processors being useful for verbose, clumsy
languages. For sure it makes sense to use them for ASM, C(++),
and sendmail, for instance.

But why for Python?
Python is short, concise, readable, has a cute class system
and introspection.
Would any of the above be improved by introducing macroes, new
syntax, namespaces, and would this cost be justified?

There are a few cases where I admit that I generate scriptlets
on-the-fly using the formatting operator. I can also imagine
that macroes could be used to make debugging easier than with
conditionals in rare cases. Maybe I would like to splice
parameterized function bodies into a caller, maybe there is
a cleaner compromise somewhere between textual replacement
and the bytecodehacks approach.

But: Plain textual replacement is covered good enough by
the % operator, if this is needed. Macroes on the token
level are fine for languages with limited namespaces, no
inheritance features and so on. Python has these to a very
high extent, so what would macroes help what cannot be done
by classes and inheritance, which justifies another level
of indirection, new syntax, and all of that?

If I would introduce macroes into Python at all, then
probably on the parser level, in order to replace nodes
by equivalent constructs which are faster or would be
otherwise less readable. 

Conclusion: 
I consider it a virtue of a language if it is powerful
enough to go without macroes. Macroes are needed for
languages which have a lack of expressiveness, except for
Scheme, where macroes are part of the language with the goal
to use them to define language constructs. If introducing
macroes for Python at all, then they should use the existing
Python structures and namespaces, macro features should be 
bound to syntactic language elements and parts of expressions,
should not introduce another level of indirection but work 
at runtime. And I'd wish to see an example where a macrofied
source file is more readable than a plain one.

IMHO, this all does not apply for Python, and it the last 
language where I would want to use macroes. 

This is no complaint against your implementation which I think
is very useful for languages which really need it.

please-keep-the-good-work - ly y'rs - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     and we keep the good language :-)



From mal at lemburg.com  Fri Aug 13 05:55:32 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 13 Aug 1999 11:55:32 +0200
Subject: ifconfig C module
References: <87g11o941w.fsf@linuxcare.com>
Message-ID: <37B3EB94.68BE201F@lemburg.com>

> I got permission to publish the results of my 'ifconfig' module.
> 
> Since it's rather small, I thought I would post the whole thing here.
> 
> I realize it's not perfect, but it does what I need.  I'll accept
> patches to make it better.  I'm sure the license could be changed to
> the Python one if there is interest in including it in the main
> distribution.

I'd suggest adding this to either the socketmodule.c or the
fcntlmodule.c in the standard Python distribution.

FYI, I've spiced up the file a bit and turned it into a package
which is available here for easier install:

	http://starship.skyport.net/~lemburg/ifconfig-0.1.zip

I changed the names a little in the above archive:

Package name: ifconfig
API:	      ifconfig.ifconfig

>From David's code:

   The command ifconfig.ifconfig(interface_name), where
   interface_name is the name of the interface, such as 'eth0' or 'lo'
   returns a dictionary with the following keys:

   hwaddr: hardware address
   addr: ip address
   brdaddr: broadcast address
   netmask: netmask

Great work, David ! I never thought it would be that easy to get
at the hardware MAC address. Now the only question remaining is:
how portable is this code ? E.g. does it compile on Windows ?

Cheers,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   140 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From aa8vb at yahoo.com  Thu Aug  5 09:40:20 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Thu, 5 Aug 1999 09:40:20 -0400
Subject: Python GUI
In-Reply-To: <7ni69d$nc7$1@nnrp1.deja.com>; from Brad on Mon, Jul 26, 1999 at 05:35:41PM +0000
References: <7ni69d$nc7$1@nnrp1.deja.com>
Message-ID: <19990805094020.A1838753@vislab.epa.gov>

Brad:
 |Is there a Python GUI builder out there anywhere?

Not that I'm aware of.

Anyone know of one in-the-works?

-- 
Randall Hopper
aa8vb at yahoo.com



From wiener at enews.nrl.navy.mil  Wed Aug 11 08:27:52 1999
From: wiener at enews.nrl.navy.mil (Zach)
Date: Wed, 11 Aug 1999 08:27:52 -0400
Subject: sys.argv attribute error
Message-ID: <37B16C47.E6212B21@enews.nrl.navy.mil>

Within a python file, I am trying to pring sys.argv, but I get an
AttributeError with argv.  I can print sys.path fine, however.  Does
anyone know why I would be having problems with printing argv?  Thanks.




From bennettt at am.appstate.edu  Fri Aug 13 16:37:52 1999
From: bennettt at am.appstate.edu (TMGB)
Date: Fri, 13 Aug 1999 16:37:52 -0400
Subject: Help: sendmail under MS NT
References: <7p1pma$ak6$1@nnrp1.deja.com>
Message-ID: <37B4821F.66E706AF@am.appstate.edu>

Being new to Python, someone else may be able to give you a more
efficient way to do this or maybe a real sendmail command.  But, using
smtplib.py I found that I had to initialize mailserver with a valid mail
server and mail recipient for it to work.  I also imported cgi, urllib,
os, and my_mailtomessages which shouldn't be needed for this snippet. 
This is running Python and Netscape Enterprise Server on an intel NT box
with the form data coming from an ALPHA NT box.


------------------
import smtplib
import string

# Get form data up here or any other code you may need

# HERE BEGINS THE MAILTO Defaults PORTION OF THE SCRIPT

   mailserver="my.mailserver.appstate.edu"
   mailrecip="me at my.mailserver.appstate.edu"
   subject="ERROR"

# Asign MAILTO results from form to mailto values   
#   first get user's domain from email address and use their server to
mail from

  
mailserver=form["sentfrom"].value[string.find(form["sentfrom"].value,"@")+1:]
   mailrecip=form["sendto"].value
   subject=form["subject"].value
   sender=form["sentfrom"].value
   message=form["message"].value
   
# send the mail
   mail=smtplib.SMTP(mailserver)
   mail.sendmail(sender,mailrecip,message)
   mail.quit

--------------------

This has not been tested with a server outside of my Domain but has
worked with different accounts on different mail servers in this domain.

Thomas

Shicheng wrote:
> 
> Under unix platforms, using, for example,
> 
> sendmail_cmd = "/usr/lib/sendmail -oi "
> 
> to access the SMTP sendmail utility.
> 
> Our enquery is: under Windows NT/95/98 platforms,
> how to access the "sendmail" utility from within
> python code?
> 
> Thanks,
> 
> Shicheng
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

-- 
----------------------------------------------------------------------
Thomas McMillan Grant Bennett           Appalachian State University
Computer Consultant II                  University Library
bennettt at am.appstate.edu               
http://www.library.appstate.edu/admin/
Voice:  828 262 6587			FAX:    828 262 3001

Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
operating system that was originally coded for a 4-bit microprocessor.
 - Chris Dunphy     Boot Magazine



From bitwise at io.com  Fri Aug 27 12:29:46 1999
From: bitwise at io.com (Bit Nik)
Date: Fri, 27 Aug 1999 11:29:46 -0500 (CDT)
Subject: A Plea For A Return To Sanity!!!
In-Reply-To: <37C68564.15207535@callware.com>
Message-ID: <Pine.LNX.4.10.9908271102120.14054-100000@eris.io.com>

Hi. 

I just recently joined this newsgroup/mailinglist and I missed out a lot
on the previous flame war (lucky me!). And I certainly don't want to
stroke the coals and recreate the flame, but i'm very curious /how/ is
Python lacking in documentation? This seems to be the major argument
against it right? Also, who are these Python biggots people are always
talking about??

I think the Python documentation is excellent, almost every question I've
had so far has been answered by searching through python.org. Sometimes
I've had trouble finding some bit of information and perhaps not even
reading all of TFM have naively posted a question here to be met with a
slew of helpful posts pointing my in the right direction. Pythoners you
are to be commended! I think I have found a language I really like, and a
part of the reason is because of all of the helpful support and good
documentation.

No doubt perl has a lot of cool things going for it too, the perl module
collection is unbeatable, perl has excellent user support even though
there are a few perl biggots. Most perlers I have met are a preaty logical
bunch, and the perl mongers are on their way to certain world domination.

Another quick question, are there such things as PUGs? (Python User
Groups). I couldn't find any information on python.org...

Andrew "mask compliments in the form of a question" Tanner




From aahz at netcom.com  Sun Aug 15 00:50:15 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Aug 1999 04:50:15 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu>
Message-ID: <7p5gu7$nlq@dfw-ixnews14.ix.netcom.com>

[c.l.perl.misc removed]

In article <37b634a2 at cs.colorado.edu>,
Tom Christiansen  <tchrist at mox.perl.com> wrote:
>
>So what do you do?  You either bail out of this untenable situation, as
>Larry Wall has done, or else "customer-service battle fatigue" sets in.
>When that happens, people end up getting snapped at, sometimes mistaking
>an honest learner for yet another beggar looking for handouts.  I dare
>say this would happen even in the best of families, given equivalent
>circumstances.  I do think that earnest individuals seriously wanting
>to work hard are still given the same helping hand they always got.

I've snipped the part of the post that I mostly agree with to disagree
somewhat with that last sentence.  Without going into details, there
were a couple of module maintainers who treated me somewhat roughly.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From urner at alumni.princeton.edu  Tue Aug 24 23:28:57 1999
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Wed, 25 Aug 1999 03:28:57 GMT
Subject: Pro Python remarks to math ed folks (pointers)
References: <37c2a51e.18050990@news.teleport.com> <Ibjw3.680$v6.61454@typhoon.nyroc.rr.com> <37c2b213.49798667@news.teleport.com> <37C30FAC.C542DB6C@cybercable.fr>
Message-ID: <37c36092.94476642@news.teleport.com>

Jean-Claude Rimbault <rimbault at cybercable.fr> wrote:

>   def __repr__(self):  # convert object to string
>      return 'Vector'+`self.data`

OK, why not?  I've added:

   def __repr__(self):
       return self.__class__.__name__ + str(self.data)

And to my Poly class:

   def __repr__(self):
       return self.__class__.__name__

>>> v  = vector.Vector(quad=(1,0,0,0))
>>> v
Vector(1, 0, 0, 0)
>>> v=v*2
>>> v=v+vector.Vector(quad=(0,1,0,2))
>>> v
Vector(2, 1, 0, 2)
>>> oTet = polys.Tetra()
>>> oTet
Tetra

Without a separate __str__ method, print has the same 
behavior as just entering the object name.  OK for my
purposes, but anyone can tweak my code.

Polyhedra have a lot of info associated with them.  My
__repr__ method might return a lot more data.  Pondering
this.

Kirby

PS:  also fixed a bug in Poly() class, which is an 
abstract class -- designed only to be subclassed and 
not used directly.

Before fix:

>>> oPoly = polys.Poly()
Traceback (innermost last):
  File "<pyshell#72>", line 1, in ?
    oPoly = polys.Poly()
  File "G:\python\polys.py", line 70, in __init__
    self.distill()  # distill edges and points from faces
  File "G:\python\polys.py", line 98, in distill
    for i in self.edges:
AttributeError: edges

After fix:

>>> oPoly = polys.Poly()
Abstract class





From fdrake at acm.org  Fri Aug 13 09:07:20 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 13 Aug 1999 13:07:20 GMT
Subject: L postfix char on longs - what is the best way to discard it.?
In-Reply-To: <934546409.21489.0.nnrp-14.d4e5b0e4@news.demon.co.uk>
References: <934546409.21489.0.nnrp-14.d4e5b0e4@news.demon.co.uk>
Message-ID: <14260.6280.851268.67080@weyr.cnri.reston.va.us>

Jim Crumpler writes:
 > query = "... WHERE guff = %s" % repr(long(x))[0:-1]
 > 
 >  -- its ugly when you use alot of longs, but does the job..

  So pack it up in a function:

	def l2a(x):
	    return `long(x)[:-1]`

	query = "... WHERE guff = %s" % l2a(x)

  It's still not pretty, but it's more clear than coding it inline
every place you need it!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From esuzm at dcs.warwick.ac.uk  Wed Aug 18 16:31:42 1999
From: esuzm at dcs.warwick.ac.uk (Andrew J Clover)
Date: 18 Aug 1999 21:31:42 +0100
Subject: Bug in ntpath.exists?
References: <7pbp9c$1b4$1@holly.csv.warwick.ac.uk> <006601bee8d1$b3b2a1f0$f29b12c2@secret.pythonware.com>
Message-ID: <7pf57e$g1c$1@stone.dcs.warwick.ac.uk>

Fredrik Lundh (fredrik at pythonware.com) wrote:

> that's an NT file system "feature".  trying to work around
> it in the interface layer is probably a bad idea (when you
> think you've got it 100% right, Microsoft releases yet an-
> other implementation...)

After spending all day tracking down other, unrelated MS bugs, I would just
like to say:

  Curse Microsoft, all their developers, and all their families!

Thank you. That's it.

>     fragment in os.listdir(prefix)

> to make sure there's really something called fragment
> with the given prefix.  (better ideas are welcome...)

Yeah, you're right. Since I'm using os.listdir to try to find similarly-
named matches, it makes sense to use it here as well. Yes! Erm, hurrah!

Cheers,

-- 
This posting was brought to you by And Clover.
(Sorry.)



From tim_one at email.msn.com  Tue Aug 17 20:04:18 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 20:04:18 -0400
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <14265.35455.590852.517197@bitdiddle.cnri.reston.va.us>
Message-ID: <000401bee90d$37627a80$eaa22299@tim>

[Jeremy Hylton]
> I've rather fond of the following approach:
>
> import random
>
> def randomize(a, b):
>     return random.choice([-1, 0, 1])
> ...
> No idea how fast this is, though not very I expect.

def test(N):
    d = {}
    for i in xrange(N * 6):
        x = range(3)
        x.sort(randomize)
        t = tuple(x)
        d[t] = d.get(t, 0) + 1
    print N, d

test(1000)

1000 {(2, 1, 0): 68,
      (0, 2, 1): 625,
      (0, 1, 2): 4443,
      (2, 0, 1): 289,
      (1, 2, 0): 155,
      (1, 0, 2): 420}

With a chi square busting distribution like that, I'm not sure speed is the
first thing you should be worried about <wink>.

a-random-decision-may-not-yield-a-random-outcome-ly y'rs  - tim





From kernr at mail.ncifcrf.gov  Fri Aug 20 01:04:40 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Fri, 20 Aug 1999 05:04:40 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
Message-ID: <37bcd7ed.45850344@news.erols.com>

I'll skip parts that have already been addressed (hopefully).

On 19 Aug 1999 08:21:55 -0700, Tom Christiansen <tchrist at mox.perl.com>
wrote:
>Per your collective requests, here is one of the documents
>with my collective observations from learning Python.

Cool.

[snips left unmarked]

>delattr(object,name)            n/a?  "delete named attribute from object".
>                                I suppose this might be
>                                        delete $obj->{name}
>                                but that's pretty uncool to do it without 
>                                asking the object.

BTW, Python does ask the object.  If object is of a builtin type, this
delattr (or "del object.attr" where "attr" is name's value) will call
the C function that the type defines for __delattr__.  If object is an
instance of a class that has a __delattr__ method, then delattr will
be equivalent to object.__delattr__(name).  If object's class doesn't
define __delattr__, then Python will simply delete that attribute.

>eval(code)                      Same as Perl, but Perl's handles
>                                statements, and python's only handles
>                                expressions.  Also, python hurls an
>                                exception if it doesn't like you,
>                                whereas Perl sets $@.  People use this
>                                to convert string to numbers.  This is
>                                a terrible security hazard.

Not used so much for converting strings to numbers since int(x),
float(x), and long(x) were changed to handle strings.  This change
(IIRC) was made from 1.4 -> 1.5, I believe, to solve this very issue.

>exec(code)                      eval(code);  plus see previous entry.

I only say this in case you revise this document for use elsewhere:
The parens are unnecessary.  "exec" is a statement keyword.

>getattr(obj,name)               n/a?  maybe just $obj->{$name}.  I don't
>                                understand why this exits.  Oh heck, yes I
>                                do.  Because you can say
>                                    name = "widget_count"
>                                    obj.name
>                                Because there's no interpolation and
>                                otherwise it will look up name.  Oh my.
>                                This is like the apply() mess.

It's really not so bad.  This actually keeps Python simple and clear.
IMO, it is not desirable for interpolation to exist for the usual
attribute access method "obj.name".  In debugging a complicated class
definition, this unambiguity can be very important.

We could add special punctuation to do this interpolation (which, if I
read you correctly, perl does?), but getattr works just as well
without futzing with the grammar.

Truly, getattr isn't used a whole lot.  I don't think that we'd
benefit from interpolation.

>hash(x)                         huh?  "returns the hash value of the object,
>                                if any"

Most (all?) immutable objects and class instances that define a
__hash__ method can be assigned a hash value with this function.  This
value is the same one that is used internally for dictionaries (IIRC).
This can be useful for some arcane things.

>input(prompt?)                  n/a.  This reads and evals the input
>                                stream!  Needed because otherwise you can't
>                                read a number.  Or you could read the
>                                string and convert it.  Ick.  Security
>                                hazard waiting to happen.

The number issue changed from 1.4 to 1.5, so no-one uses input to take
care of numbers anymore.  The only thing I think it would really be
used for anymore is a quick-and-dirty interpreter shell.

>reload(module)                  First delete from %INC, then require the
>                                module again, as in:
>                                    delete $INC{"Module.pm"}
>                                    require Module;
>                                Note that Python doesn't directly support
>                                nested modules as in
>                                    require Dir1::Dir2::Dir3::Module;
>                                except via a disgustingly named hackaround
>                                module.

Again (whoops, there goes my promise), this changed from 1.4 to 1.5.
You owe us a shrubbery.

>setattr(obj,name,value)         $obj->{$name} = $value;
>                                Interpolation wins again.  See getattr.

Shrug.  Simplicity won in Guido's mind, I guess.

>str(x)                          I thought this was the same as repr().

repr() *should* return a string that can be popped into eval and
recover the original object (well, one that is equivalent in value to
the original).  str() is sometimes used for a prettier version
designed for viewing, not evaluating.

E.g. for Numeric Python arrays
>>> from Numeric import *
>>> a = reshape(arange(16), (4,4))
>>> print repr(a)
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11],
       [12, 13, 14, 15]])
>>> print str(a)
[[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]
 [12 13 14 15]]

>tuple(seq)                      n/a.  This is a nasty hack to deal with the
>                                fact that (1,2,3) is different from [1,2,3].
>                                Hm... maybe it's like this:
>                                    $aref = [1,2,3]; 
>                                    @list = @$aref;

Not so nasty.  Tuples have properties (like immutability and
hashability) that justify their co-existence with lists.  A function
to convert objects to tuples is desirable, just like int(), float(),
list(), etc. are desirable.

>sys.exit(status)                exit(status) and no import is bloody
>                                necessary just to leave the program.
>                                But that Python does this by raising
>                                a SystemExit exception.  Perl can use
>                                END{} handlers to catch these and do
>                                at-exit processing, and Python doesn't.
>                                You have to roll your own.

Non-issue really.  There aren't many times where sys.exit() is needed.
If the interpreter is run in non-interactive mode, it will exit when
the script is completed successfully or it hits an unhandled
exception.  Exiting the interpeter from interactive mode is done by
the user with Ctrl-D on UNIX and CTRL-Z on Windows.  

Unless you really need to set the exit status, explicitly calling
sys.exit or raising SystemExit isn't usually beneficial.  The default
behaviour is sufficient for most contexts, especially debugging in
interactive mode.  Instead of exiting when it hits an unhandled
exception in interactive mode, the interpreter gives you the prompt.
>From there, you can sometimes diagnose the problem.

>sys.exitfunc                    This is one function to call on normal exit
>                                (not exception exits, unlike perl).
>                                It would have to manage all the handlers
>                                in your roll-your-own scheme.  Very ugly.

Not much of an issue, really.

>        +-------------------------------------------------------------+
>        | POSIX library.  These aren't there unless you have imported |
>        | the library module via "import posix".                      |
>        +-------------------------------------------------------------+

Generally, you should use the os module which loads the right
OS-dependent module for your environment.

Thanks for listening.

>-- 
>If you consistently take an antagonistic approach, however, people are
>going to start thinking you're from New York.   :-)
>        --Larry Wall to Dan Bernstein in <10187 at jpl-devvax.JPL.NASA.GOV>

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From rjroy at takingcontrol.com  Wed Aug 25 09:14:50 1999
From: rjroy at takingcontrol.com (Robert Roy)
Date: Wed, 25 Aug 1999 13:14:50 GMT
Subject: Python 1.5.2 on HPUX 10.20
References: <37C3D5BD.72024971@abg1.siemens.de>
Message-ID: <37c3e683.52758937@news1.on.sympatico.ca>

On Wed, 25 Aug 1999 13:38:37 +0200, Volker Hetzer
<volker.hetzer at abg1.siemens.de> wrote:

>Hi!
>This is my first look into python. I tried to install it
>% ./configure --prefix=/opt/python
>% make all install
>
>and to test it
>% make test
>
>The fcntl test failed (twice):
>
>% /opt/python/bin/python Lib/test/test_fcntl.py
>Status from fnctl with O_NONBLOCK:  0
>struct.pack:  '\000\002\000\000\000\000\000\000\000\000\000\000\000\000\000\000'
>Traceback (innermost last):
>  File "Lib/test/test_fcntl.py", line 30, in ?
>    rv = fcntl.fcntl(f.fileno(), FCNTL.F_SETLKW, lockdata)
>IOError: [Errno 22] Invalid argument
>%
>
>Is this serious? What am I to do?
>
>Greetings!
>Volker
>
>-- 
>Hi! I'm a signature virus! Copy me into your signature file to help me spread!

Hi Volker,

I have run into this problem before. Where things worked fine on our
HP-UX but not on our client's. There is a relatively simple fix.

The file lib/python1.5/plat-hp-uxB/FCNTL.py is generated from header
files. In one of the header files there is a conditional that is
evaluated. The program that generates the Python does not understand
the conditional thus defines some values twice.


....
bunch of stuff

#if !defined(_FILE64)
#  define F_GETLK	5	/* Get file lock */
#  define F_SETLK	6	/* Set file lock */
#  define F_SETLKW	7	/* Set file lock and wait */
#else
#  define F_GETLK       8	/* FILE_OFFSET_BITS=64 name space */
#  define F_SETLK       9	/* Map GET/SETLK to their 64-bit */
#  define F_SETLKW      10	/* counter parts */
#endif /* !_FILE64 */

... more stuff

This generates the following python code

# Included from sys/fcntl.h

# Included from sys/types.h

# Included from sys/_inttypes.h
MAXSUSE = 65535
MAXFUPLIM = 60000
MAXFUPLIM = 2048
FD_SETSIZE = MAXFUPLIM
FD_SETSIZE = MAXFUPLIM

# Included from .unsupp/sys/_types.h
F_DUPFD = 0
F_GETFD = 1
F_SETFD = 2
F_GETFL = 3
F_SETFL = 4
F_GETLK = 5	<<<
F_SETLK = 6	<<<
F_SETLKW = 7	<<<
F_GETLK = 8	<<<<2nd def
F_SETLK = 9	<<<<2nd def
F_SETLKW = 10	<<<< 2nd def
F_GETLK64 = 8
F_SETLK64 = 9
F_SETLKW64 = 10
F_GETOWN = 11
F_SETOWN = 12
FD_CLOEXEC = 1
F_RDLCK = 01
F_WRLCK = 02
F_UNLCK = 03
SEEK_SET = 0
SEEK_CUR = 1
SEEK_END = 2

.... and more

This means that regardless of the system under which it is compiled,
FCNTL.py sets the values of these vars to the second value. Of course
the OS expects otherwise. Thus if you have a system that uses _FILE64
your fcntl test breaks.

The fix? 3 little # marks. comment out the second set of definitions
and things should work.

Good Luck!


Robert Roy
rjroy-nospam at takingcontrol.com



From skip at mojam.com  Thu Aug  5 14:37:17 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu,  5 Aug 1999 13:37:17 -0500 (CDT)
Subject: downloading web based email
In-Reply-To: <7och4d$jr0$1@nnrp1.deja.com>
References: <7och4d$jr0$1@nnrp1.deja.com>
Message-ID: <14249.55454.471036.787883@dolphin.mojam.com>

    Preston> I'm thinking of writing a Python program to download email from
    Preston> web-based services such as Yahoo mail, Deja, Hotmail, and so
    Preston> on, and deliever the mail in a format that standard unix
    Preston> mailers can use.

    Preston> I'm posting here to make sure that this work has not already
    Preston> been done.  If someone has already seen a program like this,
    Preston> that can be adapted for a variety of web based email services,
    Preston> please let me know.  I'm not writing this as a learning
    Preston> exercise, I just need the tool. ;-)

Presuming the services you're interested in support a standard mail transfer
protocol like POP or IMAP, you need look no further than Eric Raymond's
excellent fetchmail program:

    http://www.tuxedo.org/~esr/fetchmail/

If not, perhaps you can train fetchmail to use whatever transport they do
offer. 

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From chad at vision.arc.nasa.gov  Tue Aug 24 14:40:20 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Tue, 24 Aug 1999 18:40:20 +0000
Subject: anyon know anythin about "squeeze"?
References: <7psj1a$u24h@eGroups.com> <37C1DC7D.A5B698BF@vision.arc.nasa.gov>
Message-ID: <37C2E714.8980CF01@vision.arc.nasa.gov>

Chad Netzer wrote:

> >  that there is a utility called squeeze that,
>
> This program is in the source distribution "Tools/freeze" directory;

Doh, what a bonehead mistake.  Sorry.  Others have pointed out
where "squeeze" and other neat stuff is... Sorry.

Chad





From fdrake at acm.org  Fri Aug 20 10:07:47 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 20 Aug 1999 14:07:47 GMT
Subject: Python to Perl Conversions
In-Reply-To: <1276999045-22443458@hypernet.com>
References: <1276999045-22443458@hypernet.com>
Message-ID: <14269.24883.363109.219497@weyr.cnri.reston.va.us>

Tom Christiansen wrote:
 > posix.environ                           The %ENV hash; no import
 >                                         needed in Perl.  However, assignment doesn't 
 >                                         appear to propagate to
 >                                         unborn children as it does in Perl.

Gordon McMillan writes:
 > Use putenv() to do that.


  I think the current recommendation is to use os.environ instead of
posix.environ; there's something about that in the documentation for
posix.environ as well.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From Bart.Raatgerink at cscd.nl  Tue Aug 31 09:26:54 1999
From: Bart.Raatgerink at cscd.nl (Bart Raatgerink)
Date: Tue, 31 Aug 1999 15:26:54 +0200
Subject: calling python interpreter
References: <37cbccd8.5014784@news.tiscalinet.it>
Message-ID: <37CBD81E.A96DF3FF@cscd.nl>

try +x pippo.py and then ./pippo.py


Kranio wrote:
> 
> Under Red Hat 5.x I use to call the file pippo.py as
> 
> $ python pippo.py
> 
> I would like to call it as
> 
> $ ./pippo.py
> 
> by writing the first line
> 
> #!/usr/bin/python
> 
> but it gave me "access denied" message. The interperter is in the
> right dir and I am the owner of the file; tha same happens being root.
> What is the problem?
> 
> Marco Buzzo



From bruce_dodson at bigfoot.com  Sun Aug  1 13:56:01 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Sun, 1 Aug 1999 14:56:01 -0300
Subject: NT select.select?
References: <613145F79272D211914B0020AFF6401914DC19@gandalf.digicool.com> <al8d7x7216x.fsf@sirppi.helsinki.fi>
Message-ID: <KW%o3.44971$jl.31522323@newscontent-01.sprint.ca>

It does cause a huge load, because in Winsock an fd_set is represented as an
array of SOCKETs, not as a bitmap.  So in a normal program, which uses
select for simple things like polling a single FD, a large FD_SETSIZE will
degrade performance.  For example, if FD_SETSIZE were 4096, then
select.select() would allocate about 48KB (~16KB per fd_set) on the stack
whenever it is called, even if there are just a few calls.

The good news is that, since those fake fd_sets are always passed by
reference, it is safe for the core Python to define a small FD_SETSIZE like
64 while some module overrides the builtin select with another that defines
a larger FD_SETSIZE.  Then no changes to "core" python would be required,
and only those who want the bigger fd_sets will get them.

Another thing that you can do is just dynamically allocate an array of
32-bit integers: the first holds the count, and the others hold the SOCKETs.
You can fill this without using the FD_SET macros if you are careful to
avoid duplicate FDs, or you can define FD_SETSIZE to be huge, cast the
allocated memory to fd_set *, and use the macros.  (It won't matter that
your buffer is smaller than FD_SETSIZE, because you know you're going to run
out of elements before you run out of buffer.)  If your malloc does a
reasonable job of handling small and large objects, and of re-using memory
quickly, this could be pretty much ideal.

Bruce




From philh at vision25.demon.co.uk  Sun Aug 22 19:04:25 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 22 Aug 99 23:04:25 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <37BFD9F7.7552BFB3@fnmail.com>
Message-ID: <935363065snz@vision25.demon.co.uk>

In article <37BFD9F7.7552BFB3 at fnmail.com>
           kistler at fnmail.com "Per Kistler" writes:
> Phil Hunt wrote:
> > Would anyone like to have a look at it, and possibly
> > suggest some improvements? If so, tell me, and I'll put it on
> > my website.
> 
> Let's give it a look:-) Where is it?
> -Per.

It'll put it on my website tomorrow (I'm just off to sleep now).

-- 
Phil Hunt....philh at vision25.demon.co.uk




From thehaas at my-deja.com  Sat Aug 14 13:39:51 1999
From: thehaas at my-deja.com (thehaas at my-deja.com)
Date: Sat, 14 Aug 1999 17:39:51 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu>
Message-ID: <7p49l5$uv7$1@nnrp1.deja.com>

In article <37b4ba01 at cs.colorado.edu>,
  Tom Christiansen <tchrist at mox.perl.com> wrote:
> Unbelievably stupid subject, eh?  I certainly thought so when it was
> shoved at us.  The flame war begun in comp.lang.perl.misc isn't doing
> anyone any good, and then coming over to comp.lang.python to stir up
the
> shock troops to send in a few commandos is even worse.  I don't think
> you'd like it if we in comp.lang.perl.misc sent 100,000 script kiddies
> over to comp.lang.python to rant at you guys about how K31L they
though
> Perl was and how L at M3 they thought Python was.
>

I've been reading that flame war but I noticed that it is the Perl
people, for the most part, that have donned the asbestos undies and
charged up their flame throwers (this isn't without exception, of
course).  The Python people have remained relitively calm (the original
post was cross-posted, remember?).

Although Perl and Python, as languages, are very similar, the culture
that surrounds the two languages are quite different.  I know when I've
asked someone for help in either Python or Perl, the Python people are
definitely the most helpful.  I think I can safely say, that the Perl
community is quite arrogant, especially to people new to it.  In Python,
I haven't found that to be the case at all - in fact, quite the
opposite.  They are very helpful in helping out and leading people in
the right direction.  For example: I emailed an another of one Perl
module for help in something and his reply was, "Maybe you should learn
Perl better", nevermind the fact that what I was trying to do was
undocumented.  For help with a Python module, the author told me, "The
problem here is that you are new to Python and my documentation sucks,
so here is how you do it . . ."  The funny part was, *his* documentation
was better then the Perl documentation for the module discussed above.

And, yes, I do program in both, though most often I choose Perl.
However, when I look at some of my dusty Perl programs, I think "what
the f*** was I doing?" and it takes me a while to dig through it and
figure it out.  That situation is almost impossible with Python - the
way it has to be structured makes far more readable code.   In that
respect, Python is better.

Yes, Tom, "neatness" may be subjective, but in any language, my
handwriting is terrible.  I may write in English but some may say it
looks like Arabic. However there isn't a Arab or an American that can
make heads or tails of it.  Even me sometimes . . .

>
> Please and thank you.

And thank you.
>
> --tom
>
- mikeh


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From sphilip at t2.technion.ac.il  Sat Aug  7 15:38:35 1999
From: sphilip at t2.technion.ac.il (Philip Derbeko)
Date: Sat, 07 Aug 1999 19:38:35 +0000
Subject: Help. Tracking ButtonRelease events
References: <199908071600.MAA10049@python.org>
Message-ID: <37AC8B3B.D4EBAC8A@t2.technion.ac.il>

Hello.

Thanks for your help.

Actually I wanted to reduce the complexity of this to O(1) , since search is
O(log N).
Well I've found another way of doing it , in the cost of changing a GUI a bit
:
now if you want to draw an edge you click on the source vertex ( regular click
with release)
move pointer to the destination , pointer followed by the edge , and click on
it.
This is the same model as in XFig for example.
I've done it by tracking enter and leave events for each vertex , this way i
know when and which vertex entered
and if button was clicked in it.
I'm sorry i can't post the code , its in very early development stage ,
if you want it anyway email me.

Regards
Philip

Hever Zsolt wrote:

> This is not  a direct answer, but hopefully it is a help.
> I am working on a bit similar program and I have put out some code.
> http://www.nexus.hu/hever/prog/vu/visualut.py <=main program
> http://www.nexus.hu/hever/prog/vu/frameobj.py <= module
> http://www.nexus.hu/hever/prog/vu/vu.gif    <= help, a picture of the
> program
> It grew out from the same script(s) you mentioned, there are same problems
> you have now that I had to solve somehow.
> In the frameobj.py module there is a "class Connecting_Objects:" which does
> that. I maintain an ordered taglist for additional information.
>
> Zsolt
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990807/623af36d/attachment.html>

From mhagger at blizzard.harvard.edu  Wed Aug 11 10:59:10 1999
From: mhagger at blizzard.harvard.edu (Michael Haggerty)
Date: 11 Aug 1999 10:59:10 -0400
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au>
Message-ID: <xj0vhamuzwh.fsf@cyclone.harvard.edu>

skaller at maxtal.com.au (John (Max) Skaller) writes:

> Extensions:
> 
> 	3) 'in' applies to dictionaries 'as if a sequence of keys were used'

At first I thought `hey that's cool!' since I'm kindof sick of typing
has_key() all the time.  But then I realized that this is not really
consistent with sequence types.

    item in list     IMPLIES      list[something] == item

but according to your definition,

    item in dict     IMPLIES      list[item] == something

where in each case `something' is undetermined.  I suggest you rethink
this `feature'.

Michael

-- 
Michael Haggerty
mhagger at blizzard.harvard.edu



From jgreen at RobotVision.com  Fri Aug 13 08:19:36 1999
From: jgreen at RobotVision.com (James E. Green)
Date: Fri, 13 Aug 1999 08:19:36 -0400
Subject: Does Python run under Windows 2000? (build 2031)
Message-ID: <37B40D58.DD6AADC0@RobotVision.com>

I'm running Windows 2000 RC1 (build 2031, the very latest).  Will Python
run under this OS?  Only NT4 is mentioned on the Python web page.

Thanks, Jim Green





From fredrik at pythonware.com  Wed Aug 25 08:15:39 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 25 Aug 1999 12:15:39 GMT
Subject: Newbie Needs Direction
References: <37C3BC8F.1DFFC151@megsinet.net>
Message-ID: <00be01beeef3$8bf00860$f29b12c2@secret.pythonware.com>

> I have decided to learn python and Tcl.  Are there any great tutorials
> out there for python?  I would also appreciate links to any resources
> you may know about that are full of add-on tools, etc for python.  I am
> also interested in wxPython.  I would like to also know what kinds of
> programs people are designing with python.  And your comments on the
> strengths and/or weaknesses you perceive in the language.   What
> compilers are compatible, etc.?

http://www.python.org has it all

</F>






From khalsa at acay.com.au  Thu Aug 26 06:30:43 1999
From: khalsa at acay.com.au (Gurvijay Singh Bhatti)
Date: Thu, 26 Aug 1999 20:30:43 +1000
Subject: equivalent of getch
References: <37C3C97A.59BF653E@acay.com.au> <7q0on8$h3i@news.rhrz.uni-bonn.de> <slrn7s7t0m.f0q.kaz@ashi.FootPrints.net>
Message-ID: <37C51753.620A5E3@acay.com.au>

Ok, I understand now that getch is not a C function, but still I don't know how to
wait for "enter" in JPython?

Kaz Kylheku wrote:

> On 25 Aug 1999 12:48:40 GMT, Markus Fleck <fleck at informatik.uni-bonn.de> wrote:
> >Gurvijay Singh Bhatti <khalsa at acay.com.au> wrote:
> >> could someone tell me what is equivalent of getch in jpython? I want the
> >> program to wait for the user to hit enter before going forward.
> >
> >AFAIK, this cannot be done in Java. I would
> >be surprised if it could be done in JPython.
>
> Note that waiting for enter doesn't require character-at-a-time input.  There
> is no getch() function in C, but if all you want is to wait for enter, you
> would use, say, fgets().




From garry at research.att.com  Fri Aug 27 14:44:01 1999
From: garry at research.att.com (Garrett G. Hodgson)
Date: Fri, 27 Aug 1999 18:44:01 GMT
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908271102120.14054-100000@eris.io.com>
Message-ID: <37C6DC71.447B1388@research.att.com>

Bit Nik wrote:

> I just recently joined this newsgroup/mailinglist and I missed out a lot
> on the previous flame war (lucky me!). And I certainly don't want to
> stroke the coals and recreate the flame, but i'm very curious /how/ is
> Python lacking in documentation?

the complaint seems to be about the format of the documentation.
Real Men use man pages, like God intended, not this namby-pamby HTML
crap.

-- 
Garry Hodgson			"Hey, mister, can ya tell me,
garry at sage.att.com		where a man might find a bed?"
Software Innovation Services	He just grinned and shook my hand,
AT&T Labs			"No", was all he said.



From I.Clarke at NOSPAM.strs.co.uk  Wed Aug  4 03:57:52 1999
From: I.Clarke at NOSPAM.strs.co.uk (Ian Clarke)
Date: Wed, 04 Aug 1999 08:57:52 +0100
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <7o78fb$blb$2@chief.death.de>
Message-ID: <37A7F280.1B4A8583@NOSPAM.strs.co.uk>

> How would you make a web page display this message, e.g.? I think you
> should at least define environments for Python code.

That is what the {}s are for.

>  Do you know PHP3?

I did take a look at it, and it does do something similar, but:

 A - PHP is not Python, and we all know that Python is the best language
in      the world ;-)
 B - There doesn't appear to be any facilities for web page caching
 C - It is alot larger than the system I am proposing

Ian.

-- 
Remove the "NOSPAM." from my email address before replying.



From fredrik at pythonware.com  Sun Aug 22 13:06:42 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 22 Aug 1999 17:06:42 GMT
Subject: More random python observations from a perl programmer
References: <000001beec6d$9753e240$542d2399@tim><m2iu677ujv.fsf@netmeg.net><00c601beecbc$15ac4210$f29b12c2@secret.pythonware.com> <14272.10641.952348.992882@gargle.gargle.HOWL>
Message-ID: <013e01beecc0$b5808f90$f29b12c2@secret.pythonware.com>

Les Schaffer <godzilla at netmeg.net> wrote:
> okay. but what is the significance of __this__ orthogonal in the
> statement:
> 
> "The builtins strive to stick to those type/operator combos that make
> plain sense; Guido has never been a fan of gratuitous orthogonality."
> 
> does it mean simply that guido et al believe not all type/operator
> pairs need coverage in the builtins? something like "we dont feel the
> need to span all of type/operator space since the phase space of
> python coder trajectories seems well bounded by the span of existing
> pairs"

couldn't have said it better myself ;-)

(btw, this is also related to "consistency" in the "worse-is-
better" sense: http://www.jwz.org/worse-is-better.html )

</F>




From jlj at cfdrc.com  Mon Aug 23 12:33:37 1999
From: jlj at cfdrc.com (Lyle Johnson)
Date: Mon, 23 Aug 1999 16:33:37 GMT
Subject: Python scripts for GNATS?
Message-ID: <37c17756.272463902@192.77.87.74>

Has anyone out there already written any Python scripts for use with
GNATS? I was about to go down that road since I didn't see any
references to it on the Python or GNATS web sites but thought I'd
double-check here.

What I'm specifically interested in is a script (or set of scripts)
which would compute some useful statistics on the GNATS database, e.g.
how many new bug reports this week/month versus how many closed.



From skip at mojam.com  Thu Aug 26 12:04:31 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 11:04:31 -0500 (CDT)
Subject: emacs with python mode
In-Reply-To: <37C540CF.9CE369EE@rubic.com>
References: <37C50DAD.840B67C5@xsoft.co.at>
	<37C540CF.9CE369EE@rubic.com>
Message-ID: <14277.25758.596769.209050@dolphin.mojam.com>

    Jeff> "Gerhard W. Gruber" wrote:

    >> I'm using Xemacs 21.0 Beta (August 1998) on Win32 and it works quite
    >> fine (is there a newer version out there?). Now my real
    >> question. When I'm using Python emacs recognices the mode but it
    >> doesn't automatically activate syntax highlighting,

    Jeff> I'm not very knowledgable about emacs, but I'm using
    Jeff> XEmacs on NT for Python editing.  Below are some settings
    Jeff> in my home .emacs that you might find useful.

	...

    Jeff> (setq-default indent-tabs-mode nil)

I believe this is the key setting.  If indent-tabs-mode is nil, Emacs will
only use spaces to generate indentation, so your code should work on
different machines, regardless of the size of a tab.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From skip at mojam.com  Thu Aug 26 11:56:00 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 10:56:00 -0500 (CDT)
Subject: equivalent of getch
In-Reply-To: <37C51753.620A5E3@acay.com.au>
References: <37C3C97A.59BF653E@acay.com.au>
	<7q0on8$h3i@news.rhrz.uni-bonn.de>
	<slrn7s7t0m.f0q.kaz@ashi.FootPrints.net>
	<37C51753.620A5E3@acay.com.au>
Message-ID: <14277.25469.526198.318152@dolphin.mojam.com>

    Gurvijay> Ok, I understand now that getch is not a C function, but still
    Gurvijay> I don't know how to wait for "enter" in JPython?

Try the raw_input() builtin function.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From smithy at promail.pl  Sat Aug 28 16:15:08 1999
From: smithy at promail.pl (£ukasz Kowalczyk)
Date: 28 Aug 1999 20:15:08 GMT
Subject: IDLE and iso-8859-2; threading tutorial
Message-ID: <slrn7sgenr.n2.smithy@hoohoo.org>

I was trying to use IDLE (version 0.4) on Linux but was unable to use font
with iso-8859-2 encoding. Since I'm new to Python and don't know anything
about Tk, could someone explain to me what I should be looking for in IDLE's
code?

Also, is there a tutorial about using threads in Python? I know there are a
few examples in doc/Demos directory but I would like to read about
real-world problems that may arise. I still remember how much time I spent
looking for bugs in my Java threading program and finally found out that
putting Thread.yield() all over the code did the trick...

-- 
?ukasz Kowalczyk
http://tempac.fuw.edu.pl/~lukow/



From kernr at mail.ncifcrf.gov  Thu Aug 19 18:10:38 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Thu, 19 Aug 1999 22:10:38 GMT
Subject: More random python observations from a perl programmer
References: <199908191819.MAA04074@jhereg.perl.com> <37bc71cc.19700984@news.erols.com> <37bc7de2@cs.colorado.edu>
Message-ID: <37bc7da5.22734202@news.erols.com>

On 19 Aug 1999 15:57:54 -0700, Tom Christiansen <tchrist at mox.perl.com>
wrote:

>     [courtesy cc of this posting mailed to cited author]
>
>In comp.lang.python, 
>    kernr at mail.ncifcrf.gov (Robert Kern) writes:
>:This is a holdover from the time before the
>:source and documentation were split into separate distributions.  
>
>Why was this done?  Is this a good thing?   You can't do that with Perl
>modules' documentation because all the module doc is embedded.   We did
>this to avoid a sync problem, and so people could never not have the docs.
>In fact, they can even be links to the same file, e.g. File::stat.3
>linked to and File/stat.pm.  Or do you just mean core docs, not
>module docs were split?  

The core docs (Tutorial, Ref. Manual, Ext. and Emb., etc.) plus the
Library Reference, which, I guess, *is* the module docs (except for
doc strings, which are embedded).  I think that the reasoning behind
the split was to allow Fred to update the documentation between
releases.  For the most part, this is A-Good-Thing, since the
documentation is still incomplete in some regards (probably not enough
to justify a split if we were considering the docs now, but back then
the docs were).

The issue was more of getting the documentation into sync than keeping
it there.  And, generally speaking, there aren't enough sync problems
to justify losing the flexibility of separate distributions (IMO).

>--tom
>-- 
>"IMHO, CAPS LOCK should be somewhere more
>convenient; e.g., in another building."  --jgreely

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From Cancel at Post.Gov  Sat Aug 28 05:55:39 1999
From: Cancel at Post.Gov (Cancel at Post.Gov)
Date: 28 Aug 1999 09:55:39 GMT
Subject: NEW_VER-NewsControlSystem2.1.0_EMailForDemoKey(FreeNEWSservers)
Message-ID: <7q8bmr$eqm273@www.siast.sk.ca>

  (ATT:WEBMASTER)
1.NO WAREZ OR CRACKS ON MY PAGES
2.NO SPAM COMING FROM MY SITES
3.Because Someone Doesnt like this
   doesnt justifity Deleting any Page
4.There is no ILLEGAL product or Activity on my site

WhiTeRaBBiT,s News Control System
           What Does Do???
This program does a few things:
* Allows you to forge posts (i.e. make other people look as if they
  posted something)
* Allows you to make 100% anonymous posts (when used with certain 
  NNTP servers; installed with system)
* Allows you to post to moderated groups without the approval of 
  some asshole moderator
* Allows you to cancel others posts
* Allows you to send your own "control" messages (e.g. you can remove
  groups or make your own groups)
*Includes large list of free news servers
   (search groups in free servers)Never Hunt for another Server
*Read news on one server post or cancel using another
*Mail to News posting using built in EMail Sender
   (large list of mail2news gateways)

some of the changes from 2.0.1 in 2.1.0
 Added Mail2News List to Mail2News
 Added NEW NEWS GROUP SEARCH
 Added Change in Code For Better Close

click this ling to download
http://rabbitpro.hypermart.net/
http://wowsites2.com/rabbitpro/ 

Email me at mailto:rabbitproductions at bigfoot.com
  FOR 10 DAYFREE TRIAL RUN or questions

if you are already a registered owner
  download and install over Ver2.0.0
  the program will check and install
  new serial if it matches the code
For Full Registration and UpGrades and Improvements

If you find this program usefull send
   $20.00 Money Order

Leave Pay To Blank
Include Your Registration Name:
Include Your E-Mail Address
You will Recieve Your Full Registration
By E-Mail

   Send To
 RocksU.com
 P.O. Box 468
 Zionsville,In
         46077


Thank You For Useing A
      WhiTeRaBBiT
RaBBiTproDucTionS ProDuct





(ATT:WEBMASTER)
1.NO WAREZ OR CRACKS ON MY PAGES
2.NO SPAM COMING FROM MY SITES
3.Because Someone Doesnt like this
   doesnt justifity Deleting any Page
4.There is no ILLEGAL product Activity on my site





From thomas at xs4all.nl  Tue Aug 17 02:58:25 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Tue, 17 Aug 1999 08:58:25 +0200
Subject: help printing dictionary, sorted by values.
In-Reply-To: <37B8F03F.654FA9DE@xmission.com>; from Jerry Williams on Mon, Aug 16, 1999 at 11:16:47PM -0600
References: <37B8F03F.654FA9DE@xmission.com>
Message-ID: <19990817085825.L21639@xs4all.nl>

On Mon, Aug 16, 1999 at 11:16:47PM -0600, Jerry Williams wrote:

> I have searched and searched the docs and am really frustrated with
> them.
> All I can seem to find is bits and pieces that hint about dictionaries.
> I finally found a table under mappings that listed the methods.  But
> I couldn't find anything that would list each data type and what methods
> are available.  What I am trying to do is take a file with a bunch of
> entries that appear more than once and list the count of each occurance
> in order of the highest count.  So if the file looked like:
> a
> a
> a
> b
> b
> c
> Then is would print:
> 3 a
> 2 b
> 1 c

> I thought a dictionary would work great, but can't seem to figure out
> how to sort it by values and also get the key.
> list = dict.values()
> list.sort()

> Then I get lost.  And the counts aren't going to be unique.
> Thanks for any help.  Please point me to where the methods are listed
> for
> the dictionary.  And some examples of dictionaries would be helpful.

Have you tried the Python Tutorial ?
http://www.python.org/doc/current/tut/

As for examples, here's probably what you want:

[thomas at tilburg ~]$ python
Python 1.5.1 (#1, Sep  3 1998, 22:51:17)  [GCC 2.7.2.3] on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam

>>> dict = {} 		# an empty dictionary

>>> dict['spam'] = 'eggs' # Insert a string, with a string as key
>>> dict[5.4] = [2, 3, 4] # Insert a list, with a floating point number as key
>>> dict[1,2,3] = {} 	# insert an (empty) dictionary, with a tuple as key
>>> dict['me'] = dict 	# insert a reference to the dict itself

>>> dict		# 'print' the dict
{'spam': 'eggs', (1, 2, 3): {}, 'me': {...}, 5.4: [2, 3, 4]}
>>> dict.keys()		# print the dict's keys
['spam', (1, 2, 3), 'me', 5.4]
>>> dict.values()	# print the dict's values.
['eggs', {}, {'spam': 'eggs', (1, 2, 3): {}, 'me': {...}, 5.4: [2, 3, 4]}, [2, 3, 4]]

>>> dict['spam']	# Index by string
'eggs'
>>> myindx = 5.4	
>>> dict[myindx]	# Index by variable
[2, 3, 4]
>>> dict["me"]["me"][1,2,3]	# Index-index-index
{}

>>> for key in dict.keys():	# Print all dict values by hand
...     print dict[key]
... 
eggs
{}
{'spam': 'eggs', (1, 2, 3): {}, 'me': {...}, 5.4: [2, 3, 4]}
[2, 3, 4]

What you want to do is probably something like this:

import sys

if len(sys.argv) < 2:
	print "Usage: %s <file>"%sys.argv[0]
	sys.exit()

result = {}

file = open(sys.argv[1], 'r')
for line in file.readlines():
	line = line[:-1] # We want to lose the newline
	if result.has_key(line): # This could be done with exceptions too.
		result[line] = result[line] + 1
	else:
		result[line] = 1

resultlines = result.keys()
resultlines.sort(lambda x, y: cmp(result[y], result[x]))
for line in resultlines:
	print "%d %s"%(result[line], line)

Note cmp(result[y], result[x]), not cmp(result[x], result[y]), to sort the
list highest value first. You can also do a 'resultlines.reverse()' after
the sort to get the same effect.

This is what it does:

% python mysort.py testfile
66 d
60 f
36 l
36 ;
24 u
24 k
24 n
24 j
12 e
12 i
12 h
12 2
12 0
12 r
12 w
12 v
6 ,
4 s
3 p
2 a
1 m

Nevertheless, reading the tutorial and parts of the Library Reference really
are really good ideas.

and-i-didnt-even-show-the-other-10-good-ways-to-do-it'ly y'rs.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From gherron at aw.sgi.com  Wed Aug 18 16:11:31 1999
From: gherron at aw.sgi.com (Gary Herron)
Date: Wed, 18 Aug 1999 20:11:31 GMT
Subject: Win32 Installer - new beta
References: <1278333954-16187909@hypernet.com>
Message-ID: <37BB1373.6686F844@aw.sgi.com>

Gordon McMillan wrote:
> 
>  A new beta (02) of the Win32 Installer package is available on
> starship:
> 
>  http://starship.python.net/crew/gmcm/install.html
> 
> This release incorporates bug fixes and some enhancements:
> 
>   The major changes are that command line arguments are
>   passed into your script, and a standalone / installer can be run
>   from anywhere -it is no longer assumed that the current working
>   directory is also the directory the program lives in. Files are
>   always unpacked into the "home" directory, and the absolute path of
>   the "home" directory is always available.
> 
>   See http://starship.python.net/crew/gmcm/install.html#beta02 for
>   details.

Gordon,

Nice work!  With enough work to gain an understanding of this thing, I
was able to get to within inches of what I really wanted.  I'd like to
suggest a small change to allow me to get all the way to my goal.

First the suggestion, then (in case you care) the description of what
I'd use it for.

There appears to be two differences between Run.exe and Runw.exe:
  (1)  Run.exe is a console app, and Runw.exe is not.
  (2)  Run.exe cleans up all its extracted files, and Runw.exe does not.

Are these two features necessarily connected. From looking at the
compiler switches un uplaunch.cpp, I'd say not.  What I'd really like is
a version that is NOT a console app but DOES cleanup after itself.

Is there a good reason why Runw.exe does not cleanup after itself?

OPTION 1: If not, lets compile Runw.exe with the cleanup flag on.

OPTION 2: If there is a good reason to leave Runw.exe alone, then can we
include a third Run?.exe that is compiled with the flags set the way I'm
suggesting, and perhaps, for completeness sake, a fourth version so that
all four combinations of console/no-console and cleanup/no-cleanup are
covered?

Here is why I want such thing.  I'm building a .exe (using one INSTALL
and one PYZ section) which wraps up a GUI script (pyWindows/wxWindows in
this case) with a carchive containing all of its binary dependencies and
single PYZlib containing all of its python script dependencies.  When
run, it unpacks the carchive in the local directory, and runs the script
which gets all of its binary dependencies and python imports from these
files.  Upon completion, all these files are cleaned up.  The only hitch
is the damned console window which comes up and remains blank.  (The
script is a GUI so there is no console input or output.)  If I modify
builder.py to use Runw.exe instead of Run.exe (or better yet modify the
InstallTarget class to honor the userunw flag similarly to the RunTarget
class), then the console window goes away, but the current directory is
left with a dozen or two files not cleaned up.

If you care to look at it, my config file looks like this:

[EVERYTHING]
type= INSTALL
name= app.exe
script= app.py
zlib= PYZLIB
bindepends= app.py

[PYZLIB]
name= app.pyz
dependencies= app.py, ..\installer\support\installutils.py
excludes= dospath, posixpath, macpath
includes= ..\installer\support\installutils.py

The only modification I have to make to my normal application to run in
this environment is to put
  from installutils import *
at the top and, at the bottom, execution cannot be based on the usual
  if __name__ == '__main__':

As I said before, I really like this package.  If I could get rid of the
console I'd REALLY like this package.  I may attempt to compile my own
version of Run?.exe, but I'm an old Unix weenie still learning to swim
in the deep end of the Windows pool.  Any help you could provide would
be appreciated.

Thanks
Gary Herron

-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101



From da at ski.org  Mon Aug 30 14:05:59 1999
From: da at ski.org (David Ascher)
Date: Mon, 30 Aug 1999 11:05:59 -0700 (Pacific Daylight Time)
Subject: 'if foo' calls getattr?!?
In-Reply-To: <joe-300819991050234291@chinacat.salk.edu>
Message-ID: <Pine.WNT.4.04.9908301102030.151-100000@rigoletto.ski.org>

On Mon, 30 Aug 1999, Joe Strout wrote:

> [[ This message was both posted and mailed: see
>    the "To," "Cc," and "Newsgroups" headers for details. ]]
> 
> Doing some profiling on Graphite this weekend, we found a very curious
> thing: a gazillion calls to __getattr__ on one of our classes, where we
> expected to find none.  We eventually tracked it down to a seemingly
> innocent test, basically, "if foo: self.myfoo = foo".  The "if foo" was
> invoking foo.__getattr__ (which is very expensive).
> 
> I can think of no reason why it should be this way, but it is.  Below is
> a very simple stand-alone demonstration.  It leads me to the absurd
> idiom of saying "if id(foo) != id(None)" when I really mean "if foo",
> just because using id() avoids the __getattr__ call.
> 
> Can anybody explain why this should be?  Is it a bug in Python?  Is
> there some other way to check whether I have a reference to an object
> that doesn't involve a function call?

Use 'if foo is not None' to do identity checking and not truth testing.

You're getting a getattr call because your class might define its own way
of testing truth -- for more details, see:
http://www.python.org/doc/current/ref/customization.html#l2h-134

--david







From nathan at islanddata.com  Fri Aug 20 22:23:06 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Fri, 20 Aug 1999 19:23:06 -0700 (PDT)
Subject: signals and multithreaded apps
Message-ID: <XFMail.990820192306.nathan@islanddata.com>

If I'm receiving on a socket or sleeping, even the friendliest of handled
signals will cause an IOError exception.  Thus I've put extra loops around
every sleep, receive, etc., to keep them alive.  This is a little
annoying, but doable.  However, do I also need to protect database
queries, file writes, and print statements?  Some of these are difficult
to test because they are so short, but I don't want my app to except out
on that fateful 1000th run because a SIGHUP was sent at the wrong
nanosecond that I didn't prepare for.  Any wisdom on the safes and unsafes?


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From jblake at stamp-coin.com  Sat Aug 14 19:00:10 1999
From: jblake at stamp-coin.com (Jonathon)
Date: Sat, 14 Aug 1999 23:00:10 +0000 (UTC)
Subject: ECommerce & Python
Message-ID: <Pine.LNX.4.04.9908142253380.18982-100000@natal.stamp-coin.com>


	Can somebody point me in the direction of shopping carts,
	and related items, that have been written in python.

	The only thing I have found, has been ECommerce, for Zope.
	<< An alpha release for an old version of Zope.  >>
	
	Asking, because I am about to write a shopping cart,
	and if I don't have to re-invent the wheel....

	<< And yes, I have seen umpteen shopping carts written
	in perl, none of which will correctly do what I need
	a shopping cart to do.  >>



	xan

	jonathon




From godzilla at netmeg.net  Sun Aug 22 10:28:18 1999
From: godzilla at netmeg.net (Les Schaffer)
Date: Sun, 22 Aug 1999 14:28:18 GMT
Subject: More random python observations from a perl programmer
References: <000001beec6d$9753e240$542d2399@tim>
Message-ID: <m2iu677ujv.fsf@netmeg.net>

Tim_one said to tchrist:

     > I keep thinking that everything should work on everything.

> The builtins strive to stick to those type/operator combos that make
> plain sense; Guido has never been a fan of gratuitous orthogonality.

what does orthogonality mean here? (i come from a world where (0,1)
and (1,0) are archetypal orthogonals and cant map that meaning to this
discussion)

gracias....


les schaffer



From bmcd at es.co.nz  Sun Aug 15 18:30:28 1999
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 15 Aug 1999 22:30:28 GMT
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net> <slrn7re808.77t.kaz@ashi.FootPrints.net> <7p7cel$l7h@dfw-ixnews14.ix.netcom.com>
Message-ID: <slrn7reg5o.d4.bmcd@ripley.sulaco.net>

On 15 Aug 1999 21:45:57 GMT, Aahz Maruch <aahz at netcom.com> wrote:
>Not only that, but perhaps that was the third or fourth draft of the
>code, and a while loop was more appropriate for the original draft.
>It's often the case that even when I know there's a more "appropriate"
>structure for code that I'm modifying, I leave the old code intact as
>long as there's nothing actually *wrong* with it -- fewer changes
>equates to fewer bugs (not always, but in general).

Sure, if it ain't broke ...

Unfortunately, in this case, the original specifications appear to have been
reduced, and there are huge chunks of partly implemented and even obsolete
code, including a large source source file which should have been entirely
deleted as it's functionality had been absorbed into another. This was only
deducible by examiing the linker options.

If there was any evidence of a drafting process, it wasn't obvious.

But yes, there may well have been something more complicated in the loop. For
example, if he was messing with "i" in the loop, a while loop would have been
a good way to draw another programmer's attention to the fact that something
fishy was going on, if he was so averse to comments.

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>



From mikael_aronsson at mail.bip.net  Sun Aug  8 14:22:01 1999
From: mikael_aronsson at mail.bip.net (Mikael Aronsson)
Date: Sun, 8 Aug 1999 20:22:01 +0200
Subject: Weird problem ?
Message-ID: <000101bee1ca$e9795940$46f8143e@pc58-mikael.fas.flygt.com>

Hi !

I have an application with a Python interpreter embedded and when I start it
up I load a file with PyRun_SimpleFile, the file looks like this:



# Vector class

class VECTOR:
   x = 0.0
   y = 0.0
   z = 0.0
   def __init__(self,xx=0.0,yy=0.0,zz=0.0):
      self.x = xx
      self.y = yy
      self.z = zz

# Color class

class COLOR:
    r = 0.0
    g = 0.0
    b = 0.0
    a = 0.0
    def __init__(self,rr=0.0,gg=0.0,bb=0.0,aa=1.0):
       self.r = rr
       self.g = gg
       self.b = bb
       self.a = aa

xx = VECTOR()
zz = COLOR()



This loads without any error message, but when I try to compile and run this
code:


xx = COLOR()


With PyRun_SimpleString() I get the following error:

File "<string>", line 1
xx = COLOR()
           ^
SyntaxError: invalid syntax

I can call PyRun_SimpleString() with "xx = VECTOR()", this does not give any
error, I cannot fine anything wrong with my code, does anyone have any clue
to what the problem might be ?, I am running Python 1.52 on Windows NT SP4

regards,
Mikael Aronsson






From mwh21 at cam.ac.uk  Mon Aug 30 06:37:36 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 30 Aug 1999 11:37:36 +0100
Subject: How to change default behaviour of a class?
References: <37CA5E90.28D60E87@xsoft.co.at>
Message-ID: <m3d7w58sgf.fsf@atrus.jesus.cam.ac.uk>

"Gerhard W. Gruber" <g.gruber at xsoft.co.at> writes:

> How can I change what is printed when I use the statement "print class"?
> Usually this leads to an output like <class instance at 9aad60> but I
> rather want it to print variables contained in my class instead of the
> pointer adress.

Add a __repr__ method:

class C:
    def __init__(self):
        self.a = 1
    def __repr__(self):
        return "self.a = %s"%self.a

C() => "self.a = 1" appears on stdout.

HTH,
Michael



From chad at vision.arc.nasa.gov  Fri Aug 20 20:32:39 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Sat, 21 Aug 1999 00:32:39 +0000
Subject: What is a troll by the way? [offtopic]
References: <zelv3.13444$x04.816752@typ11.nn.bcandid.com> <37bdeb1b.31915768@news.erols.com>
Message-ID: <37BDF3A7.75FEA04F@vision.arc.nasa.gov>

Robert Kern wrote:

> _Troll_ is a USENET term for someone that deliberately posts an
> inflammatory message simply to stir up argument.

I've found it funny to watch this term evolve over the past few years.
It started (I believe) as a reference to the fishing term, and used to
mean "trolling for flames" (ie. hoping to bring flames to the
discussion), perhaps employing "flame bait" for this purpose.

It is now used by many (I believe), as a derogatory reference to someone
being like an ugly, flesh-eating beast that lives under a bridge. :-)
And that calling someone a "troll" is simply meant to imply they are
contemptible for trying to incite flame wars (much like a Troll is
contemptible for, well, being a Troll).

Anyway, I personally like the first usage, and sometimes lament that
it is evolving toward the second.  I used the term "troller", rather than
"troll", but it is probably a lost cause. :-)

ob. python.  Tried "idle" for the first time today... Neat!

Chad





From nikolai.kirsebom at siemens.no  Tue Aug  3 07:24:29 1999
From: nikolai.kirsebom at siemens.no (Nikolai Kirsebom)
Date: Tue, 03 Aug 1999 11:24:29 GMT
Subject: stuck on COM interface to Word
References: <01BEDD9B.6052EFB0@exchange1.geometria.hu>
Message-ID: <37a6d12b.192256@news.mch.sni.de>

In Arpad's example on *Last print date*, the returned <time object> is
converted to an integer using the function *int*.  On my installation
I get the error: 'ValueError: illegal value'.

Should the *int* function be able to convert the <time object> and if
so - is there something I've missed out ?

Nikolai Kirsebom


On Tue, 3 Aug 1999 10:31:39 +0200, "Kiss, Arpad" <AKiss at GEOMETRIA.hu>
wrote:

>Hi Jim,
>Here is an example:
>	import time
>	print time.ctime(int(doc.BuiltInDocumentProperties("Last Print Date")))
>The online help contains other useful keywords(find the details of the
>DocumentProperty): "Title","Subject","Author" etc.
>Regards,
>Arpad
>




From cgw at fnal.gov  Tue Aug 10 17:29:22 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 10 Aug 1999 21:29:22 GMT
Subject: Getting at line numbers with parsermodule
Message-ID: <14256.39346.878530.327463@buffalo.fnal.gov>

I'm trying to work up some syntax- and consistency-checking tools, and
rather than creating my own Python parser (a la "pylint") I'm trying
to use the parser module.

It's coming along OK (modulo not having enough free time to work on
it), except I've run into one snag.  I can find certain "bad" patterns 
in the AST - one thing in particular I'm looking for is code like 
"%s %s %s" % (1,2) - but once I've identified this code I have trouble 
printing out the line number in the source where the offending code 
occurred.  I tried to simply go throught the AST object (converted to
a tuple) and look for NEWLINE tokens, count these as I go throught the 
AST to get a current line-count.  This doesn't seem to work correctly
since there are fewer NEWLINE tokens than there were lines in the
original file. 

What I'd like to be able to do is, given any node in an AST object
returned by compile.suite, is to know the line number in the input
that corresponds to this node.  Is this doable?  Any suggestions?








From gmcm at hypernet.com  Tue Aug  3 09:38:34 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 3 Aug 1999 08:38:34 -0500
Subject: Any clues to source of this delay?
In-Reply-To: <al8vhaxtg1y.fsf@sirppi.helsinki.fi>
Message-ID: <1278438765-9883643@hypernet.com>

Markus Stenberg wrote:

> Ok, as I've noted in some earlier posts, I have nasty tendency of
> prototyping some protocols in Python for occassional implementation
> in more 'robust' 'commercially feasible' languages. I've ran at some
> oddity this time, though:
> 
> Background:
> 
> The protocol involves encapsulating some data as 'packets' on TCP
> connection somewhat like the Record layer of TLSv1.  The simple
> protocol in running top of TCP; this, in itself, is nothing
> particularly new or fancy. The _problem_, however, seems to be.
> 
> Problem:
> 
> On Linux, single connections have 'glass ceiling' of 50
> roundtrips/second; no matter what I kludge, it seems to stay there.
> Only way to fix this was to send 'empty' packets after processed
> ones, thus increasing throughput by about 5X.
> 
> On NT, the glass ceiling is same, just 1/10 of Linux's (5
> roundtrips/second).
> 
> I _assume_ this is some TCP/Python-related feature; no other traffic
> than the one mentioned occurs during the test period. 

I have also run into similar (though not identical) oddities. (In my
case I was stuck with the protocol, but had to get the servers
working better). I found that Linux (AMD K6, 200Mhz) would max out
at 90 "messages" / sec, and NT (P100) at 50 / sec.

As my first cut, I wrote pure Python client and server, and ignored 
their protocol. I was maxing out the LAN connection from any box.

Then I used their protocol and their client (which can't send more 
than 40 / sec on the Linux box) and ran into these limits. Then I 
rewrote in C. The amount of CPU used dropped, but the speed did not 
increase.

So, at least in my case, there's no blame on Python. I still don't 
understand what it is about their protocol that causes this slowdown. 
It appears that select just takes longer than it "should".

- Gordon



From gregm at iname.com  Fri Aug 20 00:22:35 1999
From: gregm at iname.com (Greg McFarlane)
Date: Fri, 20 Aug 1999 14:22:35 +1000
Subject: Python to Perl Conversions
In-Reply-To: <37bc887a@cs.colorado.edu>; from Tom Christiansen on 19 Aug 1999 at 04:43:06PM
References: <37bc1303@cs.colorado.edu> <7pi0v5$nvo$2@newshost.accu.uu.nl> <37bc887a@cs.colorado.edu>
Message-ID: <19990820142235.01943@nms.otc.telstra.com.au>

     [courtesy cc of this posting mailed to cited author]

On 19 Aug, Tom Christiansen wrote:
> In comp.lang.python, 
>     m.faassen at vet.uu.nl (Martijn Faassen) writes:
> :Hm, there seems to be only a discussion on whether docs should be in man
> :pages or not in response to this posting here.
> 
> Yes, that has been disappointing.  

I also appreciated your mapping between the two languages.  How hard
would it be to write a reverse mapping?  I know very little perl, but
occasionally need to look at an existing perl script and try to figure
out what it does.  A table that gives the python equivalent of common
perl expressions could be very useful.

-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
                   of precipitation in some areas.



From smithy at promail.pl  Sun Aug 22 16:10:36 1999
From: smithy at promail.pl (£ukasz Kowalczyk)
Date: 22 Aug 1999 20:10:36 GMT
Subject: Profiler - counting individual lines' exec time
References: <slrn7rtr32.hk.smithy@hoohoo.org> <14271.17541.12909.908766@dolphin.mojam.com>
Message-ID: <slrn7s0f3i.20n.smithy@hoohoo.org>

[ Skip Montanaro ]
[...]

Thanks a lot!

-- 
?ukasz Kowalczyk
http://tempac.fuw.edu.pl/~lukow/



From bruce_dodson at bigfoot.com  Sun Aug 15 12:48:14 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Sun, 15 Aug 1999 13:48:14 -0300
Subject: Regarding the magical "or" operator
References: <37b5d712@cs.colorado.edu>
Message-ID: <1cCt3.63947$jl.40430548@newscontent-01.sprint.ca>

>My question is this: have you ever had an urge for it (or a variant)
>to care about the difference between None and any other false value?


I get along fine without it by testing "myVar is None" to distinguish it
from other false values.  However, (trying to see your point) I guess if I
were thinking about None like an IEEE NaN, or like a database NULL, it might
bother me that "not None" is 1.




From timcollinsdeja at my-deja.com  Fri Aug  6 09:52:28 1999
From: timcollinsdeja at my-deja.com (timcollinsdeja at my-deja.com)
Date: Fri, 06 Aug 1999 13:52:28 GMT
Subject: I need some advice on python
Message-ID: <7oepao$6hn$1@nnrp1.deja.com>

I jave just discovered this morning that for the foreseeable future my
work life will revolve around mobile agents (Worse things to
research....). Anyway, I want to try a simple prototype and since I
have some facility in python, I thought I'd try using it (I'd really
rather not sit around with notepad writing java aglets if I can
possibly avoid it). Problem is this :

I need a agent environment (or server or whatever it's called) that can
accept agents, that runs on Windows 98/NT. It would be a great perk if
it were cross-platform, since sooner or later Solaris and HP-UX will be
thrown into the mix, but for now Windows is the absolute neccessity.

I've seen a couple interesting projects using python for this type of
stuff, but they either run on Unix (Know-bots) or OS/2 (Rover).
Anything out there for Windows?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From I.Clarke at strs.co.uk  Wed Aug 18 04:48:11 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Wed, 18 Aug 1999 09:48:11 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com> <37B7E1AC.FF4FF2CF@strs.co.uk> <7pdoaf$b8c$1@newshost.accu.uu.nl>
Message-ID: <37BA734B.E0C6A823@strs.co.uk>

> >> The problem in comp.lang.perl.misc is not the beginners -- it is
> >> the regulars.

> Come on guys, this is not useful, not nice, and not true. Yes, of course
> people snap, and sometimes they make the mistake of not being diplomatic,
> but *everybody* makes this mistake, definitely including Ian Clarke with
> this very stupid title he chose. :)

Hey!  I was quite proud of that title ;-)  

As for the C.L.P.M regulars, I agree with the statement you quote, and
damn the consequences.  Never mind not being diplomatic, those guys are
*consistently*  arrogant, and rude.  One of their favourite tricks is to
start an argument with you and then inform you that you have been
killfiled (I have seen it happen many many times, to many many people). 
If you don't believe me have a hunt in deja-news for a few of their
utterances.  Tom Christiansen is a particularly good example, and the
situation is made worse by the way the other regulars suck up to his
every action (see "HARASSMENT -- Monthly Autoemail" thread in CLPM for a
graphic demonstration of this).

Ian.



From mal at lemburg.com  Fri Aug 20 03:45:21 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 20 Aug 1999 09:45:21 +0200
Subject: ANN: mxProxy Package - Version 0.2.0
References: <37BC70EB.7E03BBEB@lemburg.com> <7pi34c$cgg@ds2.acs.ucalgary.ca>
Message-ID: <37BD0791.978FAC9@lemburg.com>

Neil Schemenauer wrote:
> 
> M.-A. Lemburg <mal at lemburg.com> wrote:
> >2. using a weak reference which allows you to build circular
> >   reference models in your data without producing the other-
> >   wise difficult to handle memory leakage induced by this.
> 
> This is an interesting implementation of weak references.
> Thanks.  Attached is a weak reference implementation I cooked up
> yesterday.  It is more of a hack then your approach but has the
> advantage that referenced objects die sooner.  Unfortunately, my
> proxy object is incomplete.

You have an interesting implementation there too :-) Both ways
have their merrit I guess. The problem with your idea is that
you need setattr() to work on the object -- this doesn't work
on builtin types, but does do the trick for all Python instances.
My implementation builds on a true proxy that also covers
builtin types (because it also handles type slots, well, at least
most of them: the numeric slots don't work yet because of coercion
issues) and has the advantage of not modifying the referenced
object at all.

Jack Janssen just recently suggested adding a weak bit flag
to all Python objects. This would in the end cause all objects
to grow by about 2-4 bytes depending on struct alignment. With this
flag one could do much the same as suggest in your approach
(using an additional global dict like I do). However the amount
of extra storage needed is significant and probably not even
used in most cases, so the weak bit idea is controversial.

BTW, calling the checkweakrefs() function in mxProxy is pretty
much equivalent to explicitly invoking the GC in JPython.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   133 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From arcege at shore.net  Mon Aug 30 16:30:11 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 30 Aug 1999 20:30:11 GMT
Subject: Returning none
References: <000101bef12c$19566520$ad2d153f@tim> <slrn7sgrse.4ee.clgonsal@keeshah.penguinpowered.com>
Message-ID: <nRBy3.661$1k5.226217@news.shore.net>

C.Laurence Gonsalves <clgonsal at keeshah.penguinpowered.com> wrote:
: On Sat, 28 Aug 1999 04:05:31 -0400, Tim Peters <tim_one at email.msn.com> wrote:
[insert Ivan's "bobbitt/seminal" mayan glyph here]

:> > I don't think it's too hard to explicitly return None when that's
:> > what you want to do, and I would think/hope that most existing
:> > Python code does this already.
:> 
:> Python doesn't return None for the convenience of people too lazy to
:> explicitly return None themselves; it's instead a cheap way to make it
:> probable that someone using a procedure as a function will get an
:> error (None supports no operations except true/false interpretation)
:> without blowing up the interpreter (do this in C and you'll access
:> random stack trash).

: That sort of argument also says that when I go outside the bounds of a
: list in Python, I should get back None, since in C I would access random
: stack/heap trash. Instead, Python raises an exception. I think it should
: do the same if you don't return anything, and then try and access that
: nothing.

: While None does have few operations, the current behaviour does not
: guarantee a runtime error. I've very often built-up sequences filled
: with None's, or assigned None to the attributes of some object. The
: program ends up failing much later. I think it's best for Python to fail
: as close to the source of the problem as possible. Ironically, this
: "failing much later" is the same kind of behaviour you get with C
: programs that access random stack trash...

: That's why I also suggested some other possible rules that would enable
: some amount of compile-time checking, so that function writers who do
: questionable things would be alerted.

[another of Ivan's "snipping" glyphs]

: In another post I suggested another idea that requires little or no
: modifications to the grammar, but would just add some new runtime and
: compile time checks. Functions using return (or falling off the end)
: would return nothing at all (rather than None), and function definitions
: would require that the function either always return a value, or never
: returns a value. This can be very easily checked at compile time. At
: runtime, attempting to retrieve the return value of a function that
: returns nothing would raise an exception. This check could even be done
: before actually executing the function at all.

I'm guessing that most everyone has forgotten what "None" _is_.  From
Python Reference Manual: 3.2 "The standard type hierarchy":

   None
          This type has a single value. There is a single object with
          this value. This object is accessed through the built-in name
          None. It is used to signify the absence of a value in many
          situations, e.g., it is returned from functions that don't
          explicitly return anything. Its truth value is false.

The key here folks is "the absense of a value".  None means "nothing at
all" that the "procedure" proponents are all for.  It's doing exactly
what you want, you may just be thinking of it incorrectly (we all are
probably thinking of None incorrectly at times).

If you want to not include "return" in your functions, or to require
that your code must "return None" that's fine.  I think that the
argument is that the language is not going to improve by requiring
procedures.

  -Arcege




From stephen at webadmins.com  Sun Aug 29 18:10:36 1999
From: stephen at webadmins.com (Stephen)
Date: Sun, 29 Aug 1999 18:10:36 -0400 (EDT)
Subject: Why do we call python a scripting language?
In-Reply-To: <1276161941-48150991@hypernet.com>
Message-ID: <Pine.LNX.4.04.9908291810310.13853-100000@gnr.u2me3.com>

I agree 100%.

/***************************************************************\
 * Stephen ("xinu") Klassen -- Unix Systems Administrator      *
 * Website: http://stephen.webadmins.com/   UIN: 19224641 (wk) *
\***************************************************************/

On Sun, 29 Aug 1999, Gordon McMillan wrote:

> Guy Oliver wrote:
> 
> > ...Many of these people are used to thinking of scripting
> > languages as /bin/sh or batch files on DOS.  And with this mindset,
> > I dont blame them for questioning my wanting to use a 'scripting'
> > language on a project. Education is the only solution that there is,
> > but its a slow process.
> 
> I have a client like that.
> 
> I long ago decided it was easier to get forgiveness than permission.
> 
> don't-ask-don't-tell-ly y'rs
> 
> - Gordon
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 




From tim_one at email.msn.com  Tue Aug 10 23:23:59 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 10 Aug 1999 23:23:59 -0400
Subject: Lambda Functions (fwd)
In-Reply-To: <xpolakis-2708561956530001@athe530-a072.otenet.gr>
Message-ID: <000301bee3a8$f3782220$ca2d2399@tim>

> ...
> Python allows the use of lambda functions, small anonymous functions
> that can speed up processing because they are effectively inlined,

lambdas don't speed up anything in Python; they're not inlined (let alone
effectively <wink>); they're exactly as expensive as "def" functions to
invoke; they're often a little more expensive than "def" functions in
practice because of repeated construction, e.g.

def f(x):
    return x+1

for i in range(1000):
    print f(i)

constructs the function object f only once, while

for i in range(1000):
    print (lambda x: x+1)(i)

constructs an anonymous function object 1000 times.

next-thing-you-know-you'll-wake-up-in-a-tub-of-ice-with-"call-911"-
    scratched-into-your-chest-ly y'rs  - tim





From drew.mcdowell at msfc.nasa.gov  Fri Aug 27 19:00:06 1999
From: drew.mcdowell at msfc.nasa.gov (Drew McDowell)
Date: 27 Aug 1999 23:00:06 GMT
Subject: Fuzzy string matching (snippet candidate?)
References: <y0jyaextjqw.fsf@vier.idi.ntnu.no>
Message-ID: <37C718AA.544E3FB8@msfc.nasa.gov>

Magnus,
  Did you think of submitting this to the python snippets repository?
http://www.hvision.nl/~ivnowa/snippets/



"Magnus L. Hetland" wrote:
> 
> Explanation of the distance algorithm...
> 
> (Got a request for this, and thought I might as well post it...)
> 
> The algorithm:
<snip>



From tim_one at email.msn.com  Sun Aug  8 14:00:08 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 8 Aug 1999 14:00:08 -0400
Subject: map vs for (was RE: ANN: Gadfly bugfix test release)
In-Reply-To: <7oj2ep$2p2$1@nnrp1.deja.com>
Message-ID: <000801bee1c7$d9f1da40$722d2399@tim>

[Aaron Watters, times some loops]

>     for i in range(len(range1)):
>         result[i] = (range1[i], 'x')
> [vs]
>     result = map(None, range1, ['x']*len(range1))

yielding

> testing loop
> elapsed 1.10000002384
>
> testing map
> elapsed 0.330000042915

and

>     for i in range(len(range1)):
>         result[i] = range1[i]-3
> [vs]
>     from operator import sub
>     result = map(sub, range1, [3]*len(range1))

yielding

> testing loop
> elapsed 0.439999938011
>
> testing map
> elapsed 0.330000042915

Then:

> Does this mean that map is always better than for
> loops even when it looks stupid?

It's complicated.  map has only one speed advantage all the time:  it does
the loop overhead at C speed, that is without the overhead of making trips
around the Python eval loop.  Beyond that, "it depends".

map (or filter) applied to None is supernaturally fast, as your first test
showed.  Apart from zippering sequences together (or filtering out false
elements) there's no use for it, though.

Anything other than that is case-dependent.  map (or filter) usually wins if
it can be applied to something from the operator module, as your second test
showed.  But map (or filter) usually loses if applied to a Python lambda (or
function) the guts of which can be written inline when using a for-loop; in
that case map saves a bit of loop overhead at the relatively enormous
expense of making Python-level function calls the for-loop avoids.

Another complication:  in your second test, it so happens that the Python
eval loop special-cases integer "-", doing it inline.  But operator.sub goes
thru a polymorphic __sub__ interface, making a method call for each subtract
(albeit a method call all at C speed).  So the relative advantage of using
map+operator.sub vs for+"integer -" is smaller than it is for many other
operators in the operator module.  So non-None operator.something map often
has a larger advantage than your second test showed.

And one more:  if the loop trip count gets large, building up a giant
vector(s) for map (or filter) to chew over can toast your cache and start
tickling your disk.  Then it's faster to write down your problem and mail it
to c.l.py, waiting for someone else to compute the answer <wink>.

the-only-killer-advantage-is-None-ly y'rs  - tim





From reignlove at aol.com  Fri Aug  6 18:46:21 1999
From: reignlove at aol.com (reignlove at aol.com)
Date: Friday, 06 Aug 1999 16:46:21 -0600
Subject: over 18 only 85190
Message-ID: <06089916.4621@aol.com>

Adults Only:

http://207.240.225.250







cE]//^l)

{}{}{} Posted via Uncensored-News.Com, http://www.uncensored-news.com {}{}{}
{}{}{}{} Only $8.95 A Month, - The Worlds Uncensored News Source {}{}{}{}
{}{}{}{}{} Five News Servers with a BINARIES ONLY Server {}{}{}{}{}



From paul at prescod.net  Sun Aug 29 22:51:02 1999
From: paul at prescod.net (Paul Prescod)
Date: Sun, 29 Aug 1999 22:51:02 -0400
Subject: Returning none
References: <37C55044.8CA02530@prescod.net>
		<37C5683D.91D61FD0@rubic.com>
		<14277.28442.44107.591412@dolphin.mojam.com>
		<37C70821.92E810FC@prescod.net>
		<14279.9024.401900.142461@dolphin.mojam.com>
		<37C7B86E.647BA654@prescod.net> <14281.61572.191617.870559@dolphin.mojam.com>
Message-ID: <37C9F196.246EE7A7@prescod.net>

Skip Montanaro wrote:
> 
> Using make is not synonymous with programming in C.  A Makefile is nothing
> more than a record of dependencies between files and sets of actions that
> are used to create the descendant files from their ancestors when they are
> out of date.  If you have projects consisting of multiple files, you would
> probably find make (or a clone) helpful.

I know all about make, but in an all-Python project (the best kind) it
is not much value.

 Paul Prescod



From skip at mojam.com  Thu Aug 26 18:41:05 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 17:41:05 -0500 (CDT)
Subject: no leading spaces in print
In-Reply-To: <87zozechpg.fsf@kloth.slux.dyndns.org>
References: <87zozechpg.fsf@kloth.slux.dyndns.org>
Message-ID: <14277.49702.238312.729688@dolphin.mojam.com>

    Matthew> I want to print, but I don't want the leading spaces
    Matthew> currently I am using 
    Matthew> print "whatever",
    Matthew> to have multiple print lines print on a single line

    Matthew> is there any way to fix this?

Sure.  Use sys.stdout.write instead:

    sys.stdout.write("whatever")
    sys.stdout.write("morespam")
    sys.stdout.write("fewereggs")
    sys.stdout.write("\n")

will emit

    whatevermorespamfewereggs

No unasked for spaces or newlines.  You are in total control.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From phd at emerald.netskate.ru  Fri Aug  6 08:38:34 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Fri, 6 Aug 1999 16:38:34 +0400 (MSD)
Subject: Help with dictionary, please
In-Reply-To: <7oek07$so3$1@alexander.INS.CWRU.Edu>
Message-ID: <Pine.LNX.4.04.9908061635180.24989-100000@emerald.netskate.ru>

On 6 Aug 1999 rbl at hal.EPBI.CWRU.Edu wrote:
> 	 	   key = (prev2,prev1)
> 	           dict[key] = dict[key] + 1
>    and got this error:
> 	  File "Metro:Python 1.5.1:learn2.py", line 34
>           dict[key] = dict[key] + 1
>                          ^
> 	SyntaxError: invalid token
> 
> I'm obviously missing something here!

   You messed tabs and spaces. Make all things tabs or spaces, but do not
use both.

   Advice N2:
   The line
      dict[key] = dict[key] + 1
   will cause troubles, if key not in dict. You better write it the
following way:
      dict[key] = dict.get(key, 0) + 1

   dict.get is a function that returns either value for the key or default,
if the key not in the dictionary.

> Any guidance greatly appreciated!
> 
> Thank you,
> Rob Lake
> rbl at hal.cwru.edu

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From stephan at pcrm.win.tue.nl  Wed Aug 18 03:11:52 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 18 Aug 1999 09:11:52 +0200
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <14265.47748.241386.291144@buffalo.fnal.gov> <wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
Message-ID: <m3672dwojr.fsf@pcrm.win.tue.nl>

Dan Schmidt <dfan at harmonixmusic.com> writes:

> Charles G Waldman <cgw at fnal.gov> writes:
> 
> 
> Skip's method is fine if choice is allowed to equal i.  With that
> change, element exchange is no longer forced.  You can prove that
> it's a perfect shuffle by induction: the last element in the shuffled
> sequence has an equal chance of being any of the original elements,
> and so on.
> 
> The following code implements the change:
> 
>   def shuffle(lst):
>     "randomize a list"
>     for i in range(len(lst)-1,0,-1):
>       choice = int(whrandom.randint(0,i))
>       lst[choice], lst[i] = lst[i], lst[choice]

Note that this is almost equivalent to "my" algorithm (which I now
repeat for the ease of the reader):

def randomize(l): 
    length = len(l) 
    for i in range(length): 
        j = rand.randrange(i, length) 
        l[i], l[j] = l[j], l[i]

The only difference is that Skip's method, as modified by Charles,
(which makes it Charles' method, I guess) lets i walk downwards
and picks elements from the first part of the list, whereas my
method has i walking upwards and picks elements from the last part
of the list. So my method could be converted into Charles' method by
reverting the list first. But since they both randomize the list,
the original order of the list shouldn't matter.

Furthermore, Skip used he module whrandom, whereas I used random.
Any ideas in which respect these modules differ?

Finally, the use of int(...) in the 4th line of Skip's/Charles' method
seems superfluous to me.

And finally finally, perhaps a teenie weenie speed increase can still
be had by using:

def randomize(l): 
    randrange = rand.randrange
    length = len(l) 
    for i in range(length): 
        j = randrange(i, length) 
        l[i], l[j] = l[j], l[i]

,so that only local variables are used inside the inner loop.
But I didn't test whether this actually buys you anything speedwise.
And it makes the algorithm longer.

Greetings,

Stephan



From Sinistrad at penguinpowered.com  Thu Aug  5 22:05:23 1999
From: Sinistrad at penguinpowered.com (Coltrey Mather)
Date: 6 Aug 1999 02:05:23 GMT
Subject: Multi-part question: /etc/passwd and /etc/shadow   and   how to load 
 modules with a full path
Message-ID: <7odft3$got$0@199.201.191.2>

First (and mot importanlty):
    Are there any modules that could be easily incorporated into a
program to get the password for a user from /etc/shadow (e.g.: for user
authentication in a server process) that would check to see if the
password given for the username given is the right combination?


Also, if I want load a module, /home/me/mymodule.py, from a python
program, /usr/local/bin/thisprogram.py, how do I go about doing that?

Thank you.




From mwh21 at cam.ac.uk  Fri Aug 20 09:56:29 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Fri, 20 Aug 1999 14:56:29 +0100
Subject: More random python observations from a perl programmer
In-Reply-To: <14269.22704.714782.33024@weyr.cnri.reston.va.us>
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu> <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk> <14269.22704.714782.33024@weyr.cnri.reston.va.us>
Message-ID: <Pine.LNX.4.10.9908201446540.22660-100000@atrus.jesus.cam.ac.uk>

On Fri, 20 Aug 1999, Fred L. Drake, Jr. wrote:
> Michael Hudson writes:
>  > On this subject I'll just say: If people before you had cared as much
>  > about the existence of a python(1) manpage, it would have one.
> 
>   People do care about this one, and rightfully so.  Python does have
> one on Unix, and has for as long as I can remember knowing about
> Python (make your own guess as to how valuable that is).

And so it is. I didn't think there'd so much hot air about the
non-existence of something that actually exists.

Ho hum.

I even learnt about the `-d' option from it. Can't see what it does
though.

Michael




From drew.mcdowell at msfc.nasa.gov  Fri Aug 13 15:41:18 1999
From: drew.mcdowell at msfc.nasa.gov (Drew McDowell)
Date: 13 Aug 1999 19:41:18 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
Message-ID: <37B47506.14EBD269@msfc.nasa.gov>

I looked in the FAQ and couldn't find anything on the Compound
Assignment Operators.
(+=, *=, -=, etc..)
Why dosen't Python support them?  Is there a easy way to add them?

-Drew McDowell



From fdrake at acm.org  Tue Aug 10 12:36:30 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Tue, 10 Aug 1999 16:36:30 GMT
Subject: Packing and unpacking IP addresses into 32-bit integers
In-Reply-To: <NDBBIAPHEKGOBIHCBLIICEDKCAAA.mcfletch@vrtelecom.com>
References: <Pine.A32.3.90.990809163915.17213C-100000@elvis.med.Virginia.EDU>
	<NDBBIAPHEKGOBIHCBLIICEDKCAAA.mcfletch@vrtelecom.com>
Message-ID: <14256.21774.292303.186694@weyr.cnri.reston.va.us>

Mike Fletcher writes:
 > Wouldn't it be something like:
 > 	'>bbbb' or '!bbbb'
 > 
 > instead (for portability to non-big-endian machines (I'm assuming the packed
 > representation is supposed to have network byte order))?

Mike,
  There are no byte-ordering issues with byte-sized values.  Since the 
x.x.x.x is already in a specific order, the issue doesn't come up at
all here if you're only interested in coverting the dot-notation to &
from the network-byte-order binary structure.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From kaz at ashi.FootPrints.net  Mon Aug 23 17:09:32 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Mon, 23 Aug 1999 21:09:32 GMT
Subject: gcc optimization breaks NumPy?
References: <7psar7$kt1$1@nnrp1.deja.com>
Message-ID: <slrn7s3e59.8pd.kaz@ashi.FootPrints.net>

On Mon, 23 Aug 1999 20:27:34 GMT, John Fisher <jfisher at are.berkeley.edu> wrote:
>Hey folks,
>
>I was getting consistent but inexplicable segmentation faults with the
>following code, using NumPy's C API to move data between Python and C
>(for sparse matrix multiplication; the data structure is from Meschach,
>in case that's relevant).  I found the problem to be caused by gcc's
>optimization options.
>
>static PyObject * mul(PyObject *self, PyObject *args) {
>
>  ...Snip declarations...
>
>  ...Snip working code.  Function ends with:
>
>  return MakeFromMes(product);
>}
>
>which is defined as:
>
>PyObject * MakeFromMes(SPMAT *in) {
>  int i, j, m, n, nzs, count, pos, therow, thecol;
>  int dummy[0];

Ouch. This is an ANSI C constraint violation that requires a diagnostic. ANSI C
does not support zero size arrays. This code is brain-damaged.

>  double elem;
>  PyArrayObject *pr, *ir, *jc;
>
>  m = in->m; n = in->n; nzs = 0;
>
>  /* Count the number of nonzero elements of in
>     ISZERO is a macro to test for "good enough" floating point 0 */
>  for(i = 0; i < m; i++) {
>    for(j = 0; j < n; j++) {
>      if(!ISZERO(sp_get_val(in, i, j))) nzs++;
>    }
>  }
>
>  dummy[0] = nzs;

Ouch. What does this mean? Assuming that the compiler accepts the
zero length array extension (which gcc does, by default) it's an access beyond
the ``end'' of the zero-element array.

The code seems to depend on the order of allocation of auto objects in the
stack frame and on some dubious aliasing between a pseudo array object and
other auto vars.

I'm afraid that someone will have to re-write this code in the C language,
or think of a more clever hack to achieve whatever aliasing trick this is
supposed to do in somewwhat more portable manner that is more resilient against
optimization, such as maybe using a union to overlap a non-zero length array
with the other values.

>The problem occurs only with an -O option in compilation.  Other
>specific -f optimization options do not cause the seg fault.  So my

It could be that optimization causes certain of the auto vars to not be given
actual storage, but to be placed in registers. Declaring these variables as
volatile will probably prevent GCC from applying these optimizations.
Certainly, it's probably safest to use volatile when doing obtuse aliasing like
this, in case the compiler doesn't realize what is going on.

>question is -- because I've let gcc optimize other extensions to Python
>I've written without incident -- what was it here that caused this
>behavior?  I'd appreciate any ideas and speculations, because I'm still
>baffled by this, though the problem seems to be solved.

The behavior of your code is undefined. This means that a conforming C
implementation can do anything it wants. For example, bring up a friendly game
of Tetris when the offending function is called. Or make demons fly out
of your nose. :)



From barmeier at mamba.com  Sun Aug 15 07:32:55 1999
From: barmeier at mamba.com (Matthias Barmeier)
Date: Sun, 15 Aug 1999 11:32:55 GMT
Subject: Copying Huge Files ? (Newbie!)
References: <37b30e52.43353218@news.usenetserver.com>
Message-ID: <37b6a52b.330064@news.usenetserver.com>

On Thu, 12 Aug 1999 18:14:12 GMT, barmeier at mamba.com (Matthias
Barmeier) wrote:


Thanx a lot for all the answers.
I have used the shutil.py and it works fine. May a bit slow baut for
now its fast enough !

Ciao
	Matze



From skip at mojam.com  Wed Aug 25 11:15:00 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 25 Aug 1999 10:15:00 -0500 (CDT)
Subject: xanadu and python
In-Reply-To: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com>
References: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com>
Message-ID: <14276.2083.275770.273576@dolphin.mojam.com>

    Fredrik> I doubt anyone have missed that Ted Nelson's Xanadu, the
    Fredrik> original hypertext system, has finally been released as open
    Fredrik> source:

    Fredrik>     http://www.udanax.com/

    Fredrik> what Python has to do with it?  well, go to the site and find
    Fredrik> out for yourself ;-)

Well, I get:

    500 Server Error

    The hard transfer limit for this user has been reached

Not sure what that has to do with either Python or Xanadu...

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From hschroeder at gmx.net  Tue Aug 17 11:57:31 1999
From: hschroeder at gmx.net (Henning Schroeder)
Date: Tue, 17 Aug 1999 17:57:31 +0200
Subject: README: Python-URL! - the weekly guide to Python resources (Aug 16)
References: <C7374D44CCACDC61.D9C6108323B94F88.79E977ACD97F8278@lp.airnews.net>
Message-ID: <bp0cp7.dh.ln@odin.henning-schroeder.de>

David Ascher <da at ski.org> wrote:
> Python-URL is back!  
Nice to see :-) I really missed it.

Henning



From janssen at parc.xerox.com  Fri Aug  6 18:29:21 1999
From: janssen at parc.xerox.com (Bill Janssen)
Date: Fri, 6 Aug 1999 22:29:21 GMT
Subject: why are dlopen flags different between 1.5 and 1.5.2?
Message-ID: <99Aug6.152924pdt."292365"@ngsun1.parc.xerox.com>

After a fair amount of grief, I found out that while Python 1.5
specified RTLD_GLOBAL in its call to dlopen (in Python/importdl.c),
Python 1.5.2 does not.  Is there any particular reason to change this?
To be compatible with Java?

Bill



From fdrake at acm.org  Wed Aug  4 22:30:14 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 5 Aug 1999 02:30:14 GMT
Subject: Proper application of the buffer interface
Message-ID: <14248.63286.73014.204886@weyr.cnri.reston.va.us>

  There has been a bit of discussion lately regarding the buffer
interface to objects in the C API.  Some of this has been on the main
list (comp.lang.python) and some has been on other Python-related
lists.
  I have a question that I don't recall seeing answered (or asked):
What is the real intent of the buffer interface?  When should a Python 
C type implement it?
  The context is this:  the t1python package (an interface to a Type1
font renderer) has, in the past, been passing bitmaps back to Python
as string objects.  In the next release, I may create a new type in
the C layer that provides all the interesting stuff, and would like to 
be able to convert these "glyph" objects to PIL images and GTK+/GNOME
compatible images.  Does it make the most sense for the glyph objects
to offer the buffer interface to make more conversions possible
without increasing the number of memory copies, or do I misunderstand
the application of the interface?
  I'd appreciate some input on this matter.  I hope to get the next
release of t1python done before too much longer, and this is probably
the biggest remaining question that I need to deal with.
  (Any other feedback on t1python would be useful as well!)
  Thanks!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From f_beutlin at hotmail.com  Thu Aug 26 08:44:34 1999
From: f_beutlin at hotmail.com (f_beutlin at hotmail.com)
Date: Thu, 26 Aug 1999 12:44:34 GMT
Subject: Documentation string
References: <7q3c88$k8m$1@nnrp1.deja.com>
Message-ID: <7q3crf$kmo$1@nnrp1.deja.com>

In article <7q3c88$k8m$1 at nnrp1.deja.com>,
  sp00fD <sp00fD at yahoo.com> wrote:
> Can someone tell me why the
>
> def foo(self):
> """A method"""
>
> won't work?
>
> $ ./rcopy.py -p BACKUP ./foo /tmp
>   File "./rcopy.py", line 18
>     """A method"""
>                  ^
> SyntaxError: invalid syntax

The problem is simple and easily fixed: the comment is not indented. Try
indenting it, and it'll work (indentation will make the comment part of
the method or function; starting it at the same indentation level as
"def" will make it a new and incorrect command outside the function).

Python-professionals: please forgive me if I'm wrong; I'm new to Python
myself.

Regards

Joerg


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From moshez at server.python.net  Sat Aug 28 12:33:29 1999
From: moshez at server.python.net (Moshe Zadka)
Date: Sat, 28 Aug 1999 12:33:29 -0400 (EDT)
Subject: A Plea For A Return To Sanity!!!
In-Reply-To: <000301bef12c$1bfd6940$ad2d153f@tim>
Message-ID: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net>

On Sat, 28 Aug 1999, Tim Peters wrote:

> python-folk-are-the-best-buddies-psu-dues-can-buy-ly y'rs  - tim
                                   ^^^
There are a few options:

1. The mailman at Python org had a glitch
2. The timbot made a type
3. There's this underground association, Python Software Union, which goes
around breaking thumbs of people who use Python without paying the PSA.

e-mailing-cash-and-lots-of-it-ly y'rs, Z.




From tdixon.no at spam.fwi.com  Fri Aug 27 12:22:55 1999
From: tdixon.no at spam.fwi.com (Timothy Dixon)
Date: Fri, 27 Aug 1999 16:22:55 GMT
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com>
Message-ID: <37c6bb11.340373891@midpoint>

On Fri, 27 Aug 1999 10:18:55 GMT, guy_oliver at yahoo.com wrote:

>Why do we call python a scripting language?  
>
>For me, and for most of the people I try to tell about the benefits of
>python, the title 'scripting language' implies that its not a real
>language capable of real programs.  It implies to most of them that it
>is good for simple sysadmin and maybe short cgi scripts.

I put it in the class of scripting languages because it's embeddable
(as I understand it; I'm actually just a newbie) in other applications
where it can be used for scripting another app.

It's not *just* in that class, of course; polymorphism and multiple
inheritance are great real-world characteristics. :-)




From jcw at equi4.com  Sat Aug  7 10:55:14 1999
From: jcw at equi4.com (Jean-Claude Wippler)
Date: Sat, 07 Aug 1999 16:55:14 +0200
Subject: why are dlopen flags different between 1.5 and 1.5.2?
References: <99Aug6.152924pdt."292365"@ngsun1.parc.xerox.com>
Message-ID: <37AC48CA.144AEF05@equi4.com>

Bill Janssen wrote:
> 
> After a fair amount of grief, I found out that while Python 1.5
> specified RTLD_GLOBAL in its call to dlopen (in Python/importdl.c),
> Python 1.5.2 does not.  Is there any particular reason to change this?

Make that double the grief, please - it bit me too, when building the
Minotaur cross-language scripting extension.  Ended up editing Python
1.5.2 and rebuilding (and mucking with the makefile to end up with a
shared lib version of Python under Linux - but that's another thread).

Michael Hudson followed up by mentioning that RTLD_GLOBAL was discussed
here before.  If a shared lib has conflicting entry points, can't the
person who needs it write a wrapper extension which loads that offending
shlib without RTLD-GLOBAL?  I hope the onus of dealing with such a
conflict can be on the trouble-makers, not everyone else?  I haven't
looked into prior discussions, and may be missing the point completely
(yah, it happens, a lot).

-- Jean-Claude



From david.stegbauer at cz.opel.com  Tue Aug  3 10:54:55 1999
From: david.stegbauer at cz.opel.com (david.stegbauer at cz.opel.com)
Date: Tue, 3 Aug 1999 14:54:55 +0000
Subject: Annoying problem with rexec
Message-ID: <C12567C2.0056EE71.00@derumg01.cyberlink.eds.com>


Hi Ralf,

You create LOCAL name "open" and using it to open files.

Function *definition* is "executed" before run and I *guess* the local name
"open" is somehow interned. (See built-in function "intern" in library
reference.) Then during looking up for function "open" is first found interned
"open", not the built-in one. It is the same case as if you assign to "stdout"
instead of to "sys.stdout".

So you have to use either "global open" or "__builtins__.open=saved_open". The
last one is better, I thing.

David

8<----------Ralf Doering wrote:-------------
>
>def testrexec():
>        temp = "/tmp/test.1"
>        print "**** About to open:", temp
>        f2 = open(temp,"w")
>        f2.write("I WAS HERE.\n")
>        f2.close()
>        print "****",temp, "closed ."
>        saved_open = open
>        open = saved_open
>
>As a result, I get:
>
>>>> testrexec.testrexec_ok()
>**** About to open: /tmp/test.1
>Traceback (innermost last):
>  File "<stdin>", line 1, in ?
>  File "testrexec.py", line 8, in testrexec
>    f2 = open(temp,"w")
>NameError: open
>>>>
>
>
>However, if I insert a "global open" right after the start of the
>function, it works as expected.





From kzaragoza at mediaone.net  Fri Aug 27 18:37:28 1999
From: kzaragoza at mediaone.net (Kris J. Zaragoza)
Date: Fri, 27 Aug 1999 22:37:28 GMT
Subject: A Plea For A Return To Sanity!!!
References: <000001bef041$84deffa0$592d153f@tim> <37C68564.15207535@callware.com> <37C6B987.1662D46F@arlut.utexas.edu>
Message-ID: <slrn7se4ld.67s.kzaragoza@kzaragoza.ne.mediaone.net>

On Fri, 27 Aug 1999 11:15:03 -0500, Thomas Bryan <tbryan at arlut.utexas.edu> 
wrote:
>Ivan Van Laningham wrote:
>> 
>> Tim Peters wrote:
>> >
>> > Besides,
>> > Tom's very public misunderstandings are a *great* base on which to build a
>> > "Python for Perlers" FAQ.  Python isn't self-evident at first glance, and > > it's good to relive the initial confusions through his articulate eyes 
>> > <ahem> (you may not have liked how he phrased things, but there was no
>> > doubt about what he was saying -- that's good writing).
>> 
>> Quite seriously, I think this is a very good idea.  If I knew more than
>> nothing about Perl, I would volunteer to put such FAQ together; but not
>> only do I know zilch, I don't even have time to be writing this
>> letter ;-)
>
>Is anyone putting this together?  Should it be a FAQ Wizard or what?
>I could set up a FAQ Wizard on my Starship page if others are willing 
>to contribute.  I must confess that I haven't used Perl much since
>learning
>Python, so I'd need a lot of help.
>
>-------------------------------------------
>Tom Bryan
>Applied Research Laboratories
>University of Texas at Austin

If you set something up and posted a link, I would happily contribute.
I've had to quite a bit of Perl hacking (and integration with Notes!)
for an installation program on Unix.  I've seen both sides of Perl and 
Python, both as a newbie when I was learning the languages, and now as
I help others at work learn them.  With any luck, I might have some
tidbit of knowledge that will help someone else out.

-- 
Kris J. Zaragoza           | "Unfortunately, most people can't out-think a
kzaragoza at mediaone.net     | grapefruit."  --Jon Bodner



From aaron_watters at my-deja.com  Sat Aug 28 15:11:10 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Sat, 28 Aug 1999 19:11:10 GMT
Subject: Unique DB access
References: <7q0mdk$hovd@eGroups.com>
Message-ID: <7q9c8b$lom$1@nnrp1.deja.com>

In article <7q0mdk$hovd at eGroups.com>,
  cgr at formulae.co.uk wrote:
> Dear all,  (newbie)
> Can some help me?
> I have a (gadfly) database with many users data.
> The database tables continue info about confidential training details
> of all the users....

How to implement privacy, basically, via cgi...

I'd recommend this:  Use gadfly client/server mode and gadfly
server policies to prevent general purpose access to the database
(and also to simplify and speed up the cgi code).

Distinguish users via a username and password from a login screen.
If you want to be paranoid you can encrypt/timestamp the uid/pwd
and time it out and allow the user to log back in after the login
screen.  Basically the uid and password would be passed around on
the GET line as long as the user is active in the system, and if they
are inactive they must login again.

As far as the database is concerned the uid/password is just more data
in the database (not a real sql/uid) and design the policies so that
the defined accesses don't work unless the uid/passwords check out.

On the other hand you can reverse-engineer how hotmail
or egroups or similar sites do it.  They
aren't quite this paranoid, and it works pretty well I think.
The only caveat is that the user must understand that they shouldn't
leave the computer with a hotmail login active if they want privacy.

   -- Aaron Watters http://www.chordate.com

===
If god meant us to be vegetarians
why'd he make cows out of meat?
   -- seen on a bumper sticker


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jkmuir at uniserve.com  Mon Aug 30 10:01:15 1999
From: jkmuir at uniserve.com (Justin Muir)
Date: Mon, 30 Aug 1999 09:01:15 -0500
Subject: Spread/Worksheet
Message-ID: <37CA8EAB.5DD1314@uniserve.com>

Hello all.  I'm looking for a widget that has a spreadsheet(row,col)
type scrollable interface.  Anyone seen anything?


Justin.




From skip at mojam.com  Tue Aug 24 12:42:25 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 24 Aug 1999 11:42:25 -0500 (CDT)
Subject: Parrot-0.0.1
In-Reply-To: <7ps4ma$n06$1@wagner.wagner.home>
References: <935279776snz@vision25.demon.co.uk>
	<7ps4ma$n06$1@wagner.wagner.home>
Message-ID: <14274.51855.281578.240389@dolphin.mojam.com>

    Victor> Once again: who need so-called GUI builders while we have
    Victor> scripting languages, which provide same level of abstraction,
    Victor> but much more flexibilty, becouse interface is created run-time?

There are a couple reasons something like Parrot would be useful.  One, as
stated by the author, multiple backends will help give you independence from
a particular GUI toolkit.  Not everyone can or will use Tcl/Tk.  Two, even
though I could use Tcl/Tk directly, I find the Python Tkinter and PMW
interfaces much easier to use than the raw Tcl interface, mostly because you
are using a language with a more studly software engineering foundation.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From Wobbler at greenvillenc.com  Fri Aug 20 13:21:16 1999
From: Wobbler at greenvillenc.com (Daryl Zero)
Date: Fri, 20 Aug 1999 13:21:16 -0400
Subject: can some one help a newbie?
Message-ID: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com>

I was wondering what is the diffrence between python and perl and what can
you do with perl scripts and cgi on the web?





From robin at jessikat.demon.co.uk  Sun Aug 15 14:22:42 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 15 Aug 1999 19:22:42 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
 <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com>
Message-ID: <kdmYUAAyVwt3EwoS@jessikat.demon.co.uk>

In article <37b9a99b.635832158 at nntp.ix.netcom.com>, Michael Rubenstein
<miker3 at ix.netcom.com> writes
...
>The problem in comp.lang.perl.misc is not the beginners -- it is
>the regulars.
A language bigot is a language bigot, I've invented crappy laguages, I'm
just not stupid/arrogant enough to proselytise.
-- 
Robin Becker



From doughellmann at home.com  Mon Aug 30 05:32:42 1999
From: doughellmann at home.com (Doug Hellmann)
Date: Mon, 30 Aug 1999 09:32:42 GMT
Subject: running python on a floppy?
References: <z6ex3.1541$1i4.128395@typhoon.mbnet.mb.ca>
Message-ID: <37CA50EC.4FC03560@home.com>

Why are people still trying to fit distributions like this on a
*floppy*?  Why not use a Zip disk?  Granted, they are more expensive,
but there is *so* much more room.

Doug


Daniel Goertzen wrote:
> 
> Do you guys think I could squeeze a linux kernel and a python interpreter
> onto a floppy disk?  I'm concerned that the interpreter might use some
> gigantic shared libraries ...
> 
> This will be for a web accessable embedded system.  There's plenty of ram
> for ramdisks and such.
> 
> Thanks,
> Dan.



From mjackson at wc.eso.mc.xerox.com  Fri Aug 20 16:53:46 1999
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 20 Aug 1999 20:53:46 GMT
Subject: Anyone know about lgrind?
References: <y0jyaf62o1t.fsf@vier.idi.ntnu.no>
Message-ID: <7pkf8q$1tl$1@news.wrc.xerox.com>

mlh at idt.ntnu.no (Magnus L. Hetland) writes:

> Does anyone here know enough about how lgrind works to make it work
> with Python? Has anyone done this?

Never heard of it.  But the lgrindefs entry on the Beta web page
suggests that the format is common with vgrind, so try this:

# attempts to treat long strings as comments; doesn't work, but doesn't seem to hurt
Python|python|py:\
	:pb=(^\d?class\d\p\d?\:)|(^\d?def\d\p\d?\():\
	:tl:\
	:cb=#:ce=$:\
	:ab="""|''':ae="""|''':\
	:sb=":se=\e":\
	:lb=':le=\e':\
	:kw=and assert break class continue def del elif else except exec\
	finally for from global if import in is lambda not or pass print\
	raise return try while:






From MHammond at skippinet.com.au  Thu Aug  5 18:41:44 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Fri, 6 Aug 1999 08:41:44 +1000
Subject: How to link statically with python15.lib....
References: <sYdq3.157$E%3.952@news1.online.no>
Message-ID: <7od3vk$gia$1@m2.c2.telstra-mm.net.au>

You cant.  These .lib files are only import libraries for the DLL.  They do
not contain the code, just pointers to the code in the DLL.  Same goes to
Tcl/Tk.

Your only option is to completely rebuild both Python and Tcl/Tk from
sources.  I dont know about Tcl/Tk, but Python does not include any support
for that at all - you will probably need to modify the VC project files etc
to make it work.

I dont really know what you expect to save.  If you ensure you distribute
Python15.dll with your app and place it in the same directory as your app,
you shouldnt have a problem.  And Python being Python, you will have alot of
work ahead to get a true, 1 file installation.  Look at Freeze and the
"Win32 Installer" from Gordon McMillan and see if these better suit your
requirements.

Mark.

Egeland wrote in message ...
>Hello. Could someone please tell me how to link statically to python15.lib
>using MSVC6.0 Prof. Edit. ?
>I want to do this in order to avoid messing with *.dll's that may reside on
>the installation platform. I'm not
>really into install shield or anything ; just wanted to do it the easy way
>(if there is one). Guess I'll have to
>link to tcl80.lib and tk80.lib as well..?
>
>Thanks in advance,
>
>Anders Moe
>
>





From mstenber at cc.Helsinki.FI  Tue Aug  3 04:38:17 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 03 Aug 1999 11:38:17 +0300
Subject: Any clues to source of this delay?
Message-ID: <al8vhaxtg1y.fsf@sirppi.helsinki.fi>

Ok, as I've noted in some earlier posts, I have nasty tendency of
prototyping some protocols in Python for occassional implementation in more
'robust' 'commercially feasible' languages. I've ran at some oddity this
time, though:

Background:

The protocol involves encapsulating some data as 'packets' on TCP
connection somewhat like the Record layer of TLSv1.  The simple protocol in
running top of TCP; this, in itself, is nothing particularly new or
fancy. The _problem_, however, seems to be.

Problem:

On Linux, single connections have 'glass ceiling' of 50 roundtrips/second;
no matter what I kludge, it seems to stay there. Only way to fix this was
to send 'empty' packets after processed ones, thus increasing throughput by
about 5X.

On NT, the glass ceiling is same, just 1/10 of Linux's (5
roundtrips/second).

I _assume_ this is some TCP/Python-related feature; no other traffic than
the one mentioned occurs during the test period. Additionally, for pure
TCP, much higher throughput in small packets can be achieved:

P/TCP:snd+rcv                :    2110.1495/sec [1.39s] (473.90us/call)

However, with the protocol around it, the times suddenly die, literally:

Single echo(c:Up,exit:Normal)   :      23.6462/sec [4.06s] (42.290ms/call)

And the CPU spent is about 2%, thus the delays are .. somewhere. Any
insights on where? The Linux performance is sufficient, but NT performance
most definitely is _NOT_.

For reference, here's where it blocks in the Linux case:

+---------+-------------------+-------+--------------------------------------+
|Function |         Time spent|# Calls|                               Percent|
+---------+-------------------+-------+--------------------------------------+
|recv     |     79ms and 223us|    510|*                                     |
|send     |              354ms|    512|**                                    |
|select   |9s, 564ms and 750us|    530|***************95.08%***************  |
+---------+-------------------+-------+--------------------------------------+
+-------+--------------------------------------------------------------------+
|Threads|                                                            Coverage|
+-------+--------------------------------------------------------------------+
|1      |***************************90.11%****************************       |
+-------+--------------------------------------------------------------------+

Therefore it seems as if it just waits for the pending data in the select
about 90% of the time. That does not, by my definition, seem to be 'good'.

That reminds me, is there a tool for blockage coverage? CPU-use coverage is
rarely as interesting as 'where code spends it's time', to me, at any rate.

-Markus Stenberg

-- 
 Running Windows on a Pentium is like having a brand new Porsche but
 only be able to drive backwards with the handbrake on.
	(Unknown source)



From david.stegbauer at cz.opel.com  Wed Aug  4 03:47:04 1999
From: david.stegbauer at cz.opel.com (david.stegbauer at cz.opel.com)
Date: Wed, 4 Aug 1999 07:47:04 +0000
Subject: Printing from python on Windows platform
Message-ID: <C12567C3.003009B3.00@derumg01.cyberlink.eds.com>


Johann Spies write:
>In Linux I use the following to redirect printing to
>the printer:
>
>    drukker = sys.stdout
>    drukker = os.popen('lpr','w')
>    drukker.write(drukstuk)
>    sys.stdout.write(drukstuk)
>
>What would the windows equivalent be?
>
> I have tried os.popen('prn', 'w'), but without success.

Printing under windows is not such easy. I personally write to a file, then I
pass it to MS Word to print. I format the output using RTF, but I think there is
no problem to use HTML and Netscape - then it is sufficient to spawn it like

from os import spawnv, P_NOWAIT
spawnv(P_NOWAIT, r'D:\APP\NETSCAPE\PROGRAM\NETSCAPE.EXE', ('NETSCAPE.EXE',
'/print("filename.html")'))

David

8<---------------print-word.py--------------------

import struct
from time import sleep
from os import spawnv, P_NOWAIT
from dynwin import windll, windde

def PrintUsingWinWord(filename):
    app = r"C:\MSOFFICE\WINWORD\WINWORD.EXE"
    spawnv(P_NOWAIT, app, (app, "/n"))
    print 'spawned WinWord, sleeping 5 seconds'
    sleep(5)
    s = windde.dde_session()
    s.initialize()
    CC = windll.membuf(36)
    CC.write(struct.pack ('l',36) + (32 * '\000'))
    retries = 10

    while retries:
        try:
            s.connect('WinWord', 'System', CC)
            s.execute('[FileOpen("' + filename + '")]')
            #s.execute('[FilePrint 0][FileExit 2]')
            s.uninitialize()
            retries = 0
        except:
            print 'Unable to connect to WinWord, remain', retries, 'attempts'
            print 'sleeping 10 seconds'
            sleep(10)
            retries = retries - 1
    del s
    return





From jbauer at rubic.com  Thu Aug 26 12:15:57 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 26 Aug 1999 16:15:57 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net>
Message-ID: <37C5683D.91D61FD0@rubic.com>

Paul,

I think this proposal would break too much existing
code, even for a Python 2 change.

Arguably, it may not be the best coding style, but I've 
gotten used to functions returning None, if not otherwise
specified.  If your proposal could be guaranteed to 
generate a compile-time error (could it?), it might
encounter less resistance, but I still wouldn't be
very enthusiastic about the idea.  I can appreciate
the issue you're addressing, though.

Best regards,

Jeff Bauer
Rubicon Research

---
Paul Prescod wrote:
> 
> I think it would be more pythonish if this would return a runtime 
> (or even compiletime) error:
> 
> def foo():
>         a=5
> 
> b=foo()
> 
> It shouldn't just return None. That's a source of errors.
> This is obviously an incompatible change so I propose it 
> for Python 2.



From tchrist at mox.perl.com  Thu Aug 19 15:18:37 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 12:18:37 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <8E27B3FC4duncanrcpcouk@news.rmplc.co.uk>
Message-ID: <37bc4a7d@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    duncan at rcp.co.uk (Duncan Booth) writes:
:>GOTCHA: (high)
:>    Lists don't autoallocate the way dictionaries do.  So while this works
:>    on two dictionaries:
:>     new = {}
:>     for key in old.keys:
:>         new[key] = old[key]
:>    This fails on two lists:
:>     new = []
:>     for i in old:
:>         new[i] = old[i]
:This is meaningless and nothing to do with autoallocating. i is iterating 
:over values not indexes. you get the effect you want here by writing
:    	for i in old:
:    	    	new.append(i)

You're right.  I should have written

    new = []
    for i in range(0,len(old)):
	new[i] = old[i]

And that certainly is parallel to the dictionary case.  Dictionary
auto-allocate.  Lists don't.  It's a gotcha to a Perl programmer,
because he's using to having both hashes and arrays autoallocate.  
Not, of course, that we need to copy things that way, since arrays and
hashes are firstclassy. :-)  But you know what I mean.

--tom
-- 
     "... an initial underscore already conveys strong feelings of
      magicalness to a C programmer."
	--Larry Wall in <1992Nov9.195250.23584 at netlabs.com>



From spide at yahoo.com  Mon Aug 23 18:47:06 1999
From: spide at yahoo.com (spide at yahoo.com)
Date: Mon, 23 Aug 1999 22:47:06 GMT
Subject: anyon know anythin about "squeeze"?
Message-ID: <7psj1a$u24h@eGroups.com>

I saw on page 138 of Learning Python by Mark Lutz and David Ascher,
Oreilly and Associates, March 1999 that there is a utility called
squeeze that,

   "...packages Python bytecode in a Python program..."

I couldn't find anything on python.org -- anyone out there heard of it
and know where I can get it?

Thanks,
spide




From jya at ieee.org  Mon Aug 16 09:36:36 1999
From: jya at ieee.org (Jerry Avins)
Date: Mon, 16 Aug 1999 09:36:36 -0400
Subject: take the 20 second lang. survey
References: <37B4B3F0.77E6C59A@netplus.net> <GN3t3.4218$Rh.194251@typ11.nn.bcandid.com> <934619272.442851@lxms.cit.org.by>
Message-ID: <37B813E4.6DE1@ieee.org>

Eugene Zaikonnikov wrote:
> 
> David C. Hoos, Sr. <david.c.hoos.sr at ada95.com> wrote in message
> news:GN3t3.4218$Rh.194251 at typ11.nn.bcandid.com...
> >
> > Luke Loeffler <floods at netplus.net> wrote in message
> > news:37B4B3F0.77E6C59A at netplus.net...
> > > Take the 20-second language survey.  No personal information must be
> > > entered.  This is only a survey steming from my personal interest--it is
> > > not affiliated with any organization.
> > >
> > > http://www.recordingwebsite.com/language.html
> >
> >
> > I didn't answer the survey, because when it came to the questions
> > about OOL, the first internationally standardized OOL (i.e., Ada)
> > was not the choices offered.
> 
> Common Lisp/CLOS isn't listed either - although it is an ANSI standardized
> OO language. Moreover, it is not even in the 'your primary language' list.
> 
> --
>     Eugene.
When I read "Disclaimer: this survey is over. Due to inaccuracies in
polling the results are possibly inaccurate, and should be considered
so. feel free to vote for the fun of it, but a new and revised survey
should be online to
replace this one, offering more accurate results. Thank You.", I didn't
either.

Jerry 
-- 
Engineering is the art       |      Let's talk about what
of making what you want      |      you need; you may see
from things you can get.     |      how to do without it.
---------------------------------------------------------



From skip at mojam.com  Mon Aug 16 02:12:03 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 16 Aug 1999 01:12:03 -0500 (CDT)
Subject: Why use Python when we've got Perl?
In-Reply-To: <37b634a2@cs.colorado.edu>
References: <37b4ba01@cs.colorado.edu>
	<7p49l5$uv7$1@nnrp1.deja.com>
	<37b634a2@cs.colorado.edu>
Message-ID: <14263.42199.122753.147323@dolphin.mojam.com>

    Tom> There's just no way to deal with the never-ending onslaught of
    Tom> non-programmer CGI script kiddies, who seem to outnumber the rest
    Tom> of us zillions to one.

You could tell them that CGI programming is easier in Python.  In fact, the
bot that responds to first posts to c.l.p.m could do this... ;-)

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098




From phd at emerald.netskate.ru  Mon Aug 30 06:40:33 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Mon, 30 Aug 1999 14:40:33 +0400 (MSD)
Subject: How to change default behaviour of a class?
In-Reply-To: <37CA5E90.28D60E87@xsoft.co.at>
Message-ID: <Pine.LNX.4.04.9908301437480.18864-100000@emerald.netskate.ru>

On Mon, 30 Aug 1999, Gerhard W. Gruber wrote:
> How can I change what is printed when I use the statement "print class"?
> Usually this leads to an output like <class instance at 9aad60> but I
> rather want it to print variables contained in my class instead of the
> pointer adress.

   "print" uses str() to represent an object. You can override str()
representation by providing __str__ or __repr__ methods in your calss. If
you provide both, you can define different representations for str() and
repr() functions. If you provide obly one, it is good idea to set other:

   __repr__ = __str__ # if you define __str__

> -- 
> bye,
>    Gerhard
> email: sparhawk at eunet.at
>        g.gruber at xsoft.co.at

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From ngps at madcap.dyn.ml.org  Fri Aug 27 11:28:52 1999
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 27 Aug 99 15:28:52 GMT
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com> <001f01bef078$25fcf870$f29b12c2@secret.pythonware.com> <xlxk8qhs3y6.fsf@gold.cis.ohio-state.edu>
Message-ID: <37c6aeb4.0@news.cyberway.com.sg>

According to Matt Curtin  <cmcurtin at interhack.net>:
> I rather like Larry Wall's definition of "script": something that you
> write to accomplish a quick job at hand, as opposed to a "program":
> something you write with the intention of putting it into some sort of
> production.

I rather like Larry Wall's other definitions: "script" is something
you give the actors, "program" is something you give the audience. ;-)



-- 
Ng Pheng Siong <ngps at post1.com>




From fdrake at acm.org  Fri Aug 20 10:17:39 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 20 Aug 1999 14:17:39 GMT
Subject: Python to Perl Conversions
In-Reply-To: <37BCDA12.6CB5B89A@Lugoj.Com>
References: <37bc1303@cs.colorado.edu>
	<37BCDA12.6CB5B89A@Lugoj.Com>
Message-ID: <14269.25475.342876.556750@weyr.cnri.reston.va.us>

James Logajan writes:
 > the STL. For your benefit I've included the Python man page as an
 > attachment. I hope this helps clear up any misunderstanding you may have
 > about Python documentation.
...
 > .TH PYTHON "7 September, 1997"

James,
  This is old.  I'll be glad to send anyone who asks (nicely, now!) a
current version, or (perhaps better) put it on the Web site if there
are enough requests.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From judi at user2.infinet.com  Sun Aug 15 14:27:27 1999
From: judi at user2.infinet.com (Judi Nielsen)
Date: 15 Aug 1999 14:27:27 -0400
Subject: python and win32all
Message-ID: <7p70qf$ndg@user2.infinet.com>

I'm thinking of using zope as a web application server and was 
wondering . .

Zope installs python for you, can one also install the win32all
software with it and use it with zope?

Thanks for any info,

john

(p.s. sorry if this is a duplicate my other news server doesn't seem to
want to post)





From cbbrowne at news.hex.net  Sat Aug 28 14:07:29 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: 28 Aug 1999 18:07:29 GMT
Subject: UML
References: <3F8BA119B48AD11197470008C7A493CB98B502@HQOBGEX1> <37c2afa8.67206110@news.compuserve.com> <slrn7s69p4.jb7.wtanksle@dolphin.openprojects.net> <37C3D89C.D57E9956@abg1.siemens.de> <935618487snz@vision25.demon.co.uk> <37C6BD40.4FA0640@gssec.bt.co.uk>
Message-ID: <7q98h1$n285@george.sabre.com>

On Fri, 27 Aug 1999 17:30:56 +0100, Alan Gauld
<alan.gauld at gssec.bt.co.uk> wrote: 
>> > Nutshell books are reference books. I tried that one as well and was
>> > disappointed too.
>> 
>> I didn't like _UML in a Nutshell_ either, it's the worst O'Reilly
>> book I've ever read.
>
>Its not even a good reference! It misses several bits out
>(cf coverage of Activity diagrams)

I have not yet heard any good report on that book.

>I prefer Instant UML from Wrox press - but it does have 
>a slightly limited view of UML, and tries to explain UML 
>in terms of UML which may not suit everyone!
>
>Fowler's book gets the best reviews so far.

The Instant UML book has a (seemingly unique) section on mapping UML
to IDL, which strikes me as being one way in which to make sure that
it is actually useful.

Automating a particular diagramming notation isn't that useful if
you're still left with having to write up manually how it corresponds
to the actual code.

As near as I can tell, the real value in having UML would be in
diagramming the relationships between the objects whose formal
interfaces would be outlined in IDL.  

There don't seem to be many tools yet that provide that
correspondence, perhaps unfortunately...

-- 
A student, in hopes of understanding the Lambda-nature, came to Greenblatt.
As they spoke a Multics system hacker walked by.  "Is it true", asked the
student, "that PL-1 has many of the same data types as Lisp?"  Almost before
the student had finished his question, Greenblatt shouted, "FOO!", and hit
the student with a stick.
cbbrowne at ntlug.org- <http://www.hex.net/~cbbrowne/lsf.html>



From claird at starbase.neosoft.com  Thu Aug 26 15:46:50 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Aug 1999 19:46:50 GMT
Subject: fully qualified hostname ?
References: <199908261822.AA01099@ch1d2833nwk>
Message-ID: <595C6D6DD3292BB3.DCF1BAEEAE3E421A.EFDF99C4B418A66D@lp.airnews.net>

In article <199908261822.AA01099 at ch1d2833nwk>,
Ted Horst  <Ted.Horst at wdr.com> wrote:
>
>What is the best way to get the fully qualified hostname in python (ie  
>hostname.domainname) ?
>
>A portable solution would be ideal, but I would settle for Unix only.
			.
			.
			.
There's a lot of ambiguity in the question of which
you're not aware.  A perfectly well-behaved network
host might have several fully-qualified hostnames,
or none.  One can even make a case that a single
host is best known by different names on different
ports.

So:  you're definitely entering the land of some-
what sloppy heuristics.  The first thought that
occurs to me is a reverse DNS lookup; is that ade-
quate to your situation?

Note, also, that many Python instances run on host-
nameless hosts.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From markh at usai.asiainfo.com  Tue Aug 31 05:31:04 1999
From: markh at usai.asiainfo.com (markh at usai.asiainfo.com)
Date: Tue, 31 Aug 1999 09:31:04 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <935457373snz@vision25.demon.co.uk>
Message-ID: <gr6gq7.jd.ln@bigbird.sw.asiainfo.com>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
: In article <7ps4ma$n06$1 at wagner.wagner.home>
:            vitus at wagner.rinet.ru "Victor Wagner" writes:

:> Really it is very simple to define parrot interpreter on tcl - just
:> define commands such as window, columnLayout, rowLayout as tcl procedures 
:> which substitute
:> something by default.

: Would anyone want to do this in Tcl as opposed to Python? (Not
: wishing to get into another language war :-)).

Interestingly, your parrot specifications follow Tcl's syntax,
so it's very simple to make a complete parrot parser in tcl.
Stick these lines in  front of your parrot file and run it
through tclsh.  Modify the "puts" commands appropriately
and you will have the front end you need.

-------------- cut here ----------------
# parrot parser
proc window {name title cmds} {
    puts "window: $name"
    eval $cmds
}
proc menubar {cmds} {
    puts "menubar"
    eval $cmds
}
proc menu {name cmds} {
    puts "menu: $name"
    eval $cmds
}
proc menuBar {cmds} {
    puts "menuBar"
    eval $cmds
}
proc menuItem {name label} {
    puts "menuItem: $name"
}
proc colLayout {cmds} {
    puts "colLayout"
    eval $cmds
}
proc rowLayout {cmds} {
    puts "rowLayout"
    eval $cmds
}
proc label {label} {
    puts "label: $label"
}
proc button {name label} {
    puts "button: $name"
}
proc radioButton {name label} {
    puts "radioButton: $name"
}
-------------- cut here ----------------

-- 
Mark Harrison                       "Open the floppy disk door, Hal."
AsiaInfo Computer Networks          http://usai.asiainfo.com:8080/
Beijing, China / Santa Clara, CA    markh at usai.asiainfo.com



From fredrik at pythonware.com  Thu Aug 19 06:00:57 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 19 Aug 1999 10:00:57 GMT
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org> <00c301bee884$42b1d3b0$f29b12c2@secret.pythonware.com> <37BB8B9B.6092BB93@Lugoj.Com>
Message-ID: <001101beea29$bf2c68f0$f29b12c2@secret.pythonware.com>

James Logajan <JamesL at Lugoj.Com> wrote:
> Fredrik Lundh wrote:
> > This module [os] provides a more portable way of using
> > operating system (OS) dependent functionality than
> > importing an OS dependent built-in module like posix
> > or nt.

note that I didn't really write that -- I just quoted
the handbook.

> Only MS would fix something that wasn't broken. But when there is a
> world to conquer.... I vaguely recall once hearing Bill Gates saying Unix
> was going to be the OS of the future. This was back when MS pushed
> Xenix as THE robust OS; before DOS really came on the scene.)

I recall reading somewhere that back in the old days,
Microsoft did most development on Unix/68k boxes of
their own design.  that was before a bunch of guys
decided to start a company to sell the Stanford
University Network computer to a wider audience.
things could have been quite a bit different...

> This was back when MS pushed Xenix as THE robust OS.

Xenix was quite cool.  we ran multiuser system on
8 MHz 286 boxes in the mid-eighties.  much faster
than the Z80 systems our customers had used be-
fore that ;-)

</F>




From adid543121 at aol.com  Tue Aug 24 14:28:20 1999
From: adid543121 at aol.com (adid543121 at aol.com)
Date: Tue, 24 Aug 1999 12:28:20 -0600
Subject: $1/MINUTE PHONESEX W/ REAL PEOPLE LIKE YOURSELF 83982
Message-ID: <240899122820@aol.com>

Wanna meet someone tonight?

It costs ONLY $1 per minute

CALL NOW :  1-800-750-GIRL (4475)

These girls are not phonesex workers. They are horny girls who are on the line cuz it's free phonesex for them. You are free to arrange to meet anyone you meet on this line. Give it a try!

-Enjoy


oBo)x>9`



From phil at river-bank.demon.co.uk  Tue Aug 31 14:08:18 1999
From: phil at river-bank.demon.co.uk (Phil Thompson)
Date: Tue, 31 Aug 1999 19:08:18 +0100
Subject: PyKDE for Win ?
References: <7q1als$id9$1@ash.prod.itd.earthlink.net>
Message-ID: <936122968.500204282@news.demon.co.uk>

On Wed, 25 Aug 1999, Kikutani Makoto wrote:
>Looks like a ridiculous qestion, but is there any plan to
>port PyKDE to Win world ?
>
>-- 
>Kikutani, Makoto  kikutani at sprintmail.com

It's on the todo list (at least a port of the Qt part of the bindings), but not
very high up on that list at the moment.

Phil



From psoares at consiste.pt  Wed Aug  4 06:36:19 1999
From: psoares at consiste.pt (Paulo Soares)
Date: Wed, 4 Aug 1999 10:36:19 GMT
Subject: Printing from python on Windows platform
Message-ID: <51998F46F8B7D011BF7C0060B03502AD0EA9DB@orion.consiste.pt>

I suppose that you are trying to send bytes directly to the printer,
by-passing any windows printer driver. I use the following code (from
C/C++) and if I have the time I'll put it into a python extension. Note
that this code is NOT thread safe and was done to allow Visual Basic to
send data to the printer.


enum {OPEN_PRINTER, WRITE_PRINTER, CLOSE_PRINTER, ABORT_PRINTER};

extern "C"
__declspec(dllexport) long __stdcall drtprint(long action, LPCSTR text,
long length)
{
    static HANDLE hPrinter = NULL;
    switch (action)
    {
    case OPEN_PRINTER:
        if (hPrinter)
            return -1;
        if (OpenPrinter((char *)text, &hPrinter, NULL))
        {
            DOC_INFO_1 Info;
            PRINTER_INFO_2 *pp2 = NULL;
            DWORD Wrn;
            GetPrinter(hPrinter, 2, (LPBYTE)pp2, 0, &Wrn);
            BYTE *pp = new BYTE[Wrn + 1];
            GetPrinter(hPrinter, 2, (LPBYTE)pp, Wrn + 1, &Wrn);
            pp2 = (PRINTER_INFO_2 *)pp;
            char    DocName[128], OutputFile[128], DataType[128];
            memset(&Info, 0, sizeof(Info));
            strcpy(DocName, "Impress?o");
            strcpy(OutputFile, pp2->pPortName);
            strcpy(DataType, "");
            Info.pDocName       = DocName;
            Info.pOutputFile    = OutputFile;
            Info.pDatatype      = 0;
            delete [] pp;
            if (!StartDocPrinter(hPrinter, 1, (LPBYTE)&Info))
            {
                DWORD err = GetLastError();
                ClosePrinter(hPrinter);
                hPrinter = NULL;
                return err;
            }
            return 0;
        }
        else
        {
            DWORD err = GetLastError();
            hPrinter = NULL;
            return err;
        }
        break;
    case WRITE_PRINTER:
        if (!hPrinter)
            return -1;
        {
            DWORD Written;
            if (!WritePrinter(hPrinter, (void *)text, length, &Written))
            {
                DWORD err = GetLastError();
                ClosePrinter(hPrinter);
                hPrinter = NULL;
                return err;
            }
            return 0;
        }
        break;
    case CLOSE_PRINTER:
        if (!hPrinter)
            return -1;
        {
            DWORD err = 0;
            if (!EndDocPrinter(hPrinter))
            {
                err = GetLastError();
            }
            ClosePrinter(hPrinter);
            hPrinter = NULL;
            return err;
        }
        break;
    case ABORT_PRINTER:
        if (!hPrinter)
            return -1;
        {
            DWORD err = 0;
            if (!AbortPrinter(hPrinter))
            {
                err = GetLastError();
            }
            ClosePrinter(hPrinter);
            hPrinter = NULL;
            return err;
        }
        break;
    }
    return -1;
}

Best Regards,
Paulo Soares

> -----Original Message-----
> From:	Johann Spies [SMTP:jhspies at futurenet.co.za]
> Sent:	Tuesday, August 03, 1999 22:26
> To:	Python-poslys
> Subject:	Printing from python on Windows platform
> 
> I am sorry to ask this a second time, but I did not receive any hint 
> after the first time I put this question to the list. 
> 
> I am a Linux user writing a program for somebody that will use it on a
> Windows platform.  In Linux I use the following to redirect printing
> to
> the printer:
> 
>     drukker = sys.stdout
>     drukker = os.popen('lpr','w')
>     drukker.write(drukstuk)
>     sys.stdout.write(drukstuk)
> 
> What would the windows equivalent be?
> 
> I have tried os.popen('prn', 'w'), but without success.  I cannot see
> anything in the python documentation to help me. 
> 
> 
> Johann
> 
> 
> ----------------------------------------------------------------------
> ----
> | Johann Spies                                 Windsorlaan 19
> |
> | jhspies at futurenet.co.za                3201 Pietermaritzburg
> |
> | Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika
> (South Africa)  |
> 
> ----------------------------------------------------------------------
> ----
> 
>      "Have not I commanded thee? Be strong and of a good 
>       courage; be not afraid, neither be thou dismayed; for 
>       the LORD thy God is with thee whithersoever thou 
>       goest."                        Joshua 1:9 
> 
> 



From pinard at iro.umontreal.ca  Tue Aug 31 23:21:36 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: Wed, 1 Sep 1999 03:21:36 GMT
Subject: Turing Compliant?
In-Reply-To: Christian Tismer's message of "Thu, 02 Sep 1999 02:57:11 +0200"
References: <37CD7CBC.54AD295F@megsinet.net> <14285.34671.165473.168931@dolphin.mojam.com> <oqpv02cq8v.fsf@titan.progiciels-bpi.ca> <37CDCB67.1A3C097C@appliedbiometrics.com>
Message-ID: <oqiu5vz58f.fsf@titan.progiciels-bpi.ca>

Christian Tismer <tismer at appliedbiometrics.com> ?crit:

> Well, this is not so hard: Emulate the "infinite tape" by a class which
> is able to read and write from removeable disks, and whenever a partial
> tape end is hit, let the user "insert a new disk" on demand.
> Which is of course limited by money.

And also, by the number of elementary particles in the whole universe.
That is, not enough for Python to merely hold the value of Ackerman(4, 4).
But a Turing machine can "easily" compute that number.

Unless I miss something basic, Turing completeness considerations for
programming languages look, how should I put it, _slightly_ pedantic. :-)

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From kernr at mail.ncifcrf.gov  Tue Aug 17 17:58:33 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Tue, 17 Aug 1999 21:58:33 GMT
Subject: Error Message: Can't assign to Operator
References: <Pine.LNX.4.04.9908172133450.24702-100000@natal.stamp-coin.com>
Message-ID: <37b9da9f.1298820@news.erols.com>

[posted and e-mailed]

On Tue, 17 Aug 1999 21:41:01 +0000 (UTC), Jonathon
<jblake at stamp-coin.com> wrote:
>
>	I keep hitting this error message:
>	SyntaxError:  can't assign to operator(line 3056)
>	
>	Line 3056 is
>	meta-previous = string.join( the_ultimate_file_destination_is_this, string_record_id, suffix )
            ^
There's your problem.  You can't include a '-' in the middle of a
name.  Change 'meta-previous' to something like 'meta_previous'.

[snip]

>	jonathon

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From gerrit.holl at pobox.com  Sun Aug 29 03:20:44 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Sun, 29 Aug 1999 07:20:44 GMT
Subject: echoing input
In-Reply-To: <7q92j8$naa$1@demon.uunet.ca>; from Greg Wilson on Sat, Aug 28, 1999 at 12:24:32PM -0400
References: <7q92j8$naa$1@demon.uunet.ca>
Message-ID: <19990829092044.A5809@optiplex.palga.uucp>

On Sat, Aug 28, 1999 at 12:24:32PM -0400, Greg Wilson wrote:
> Hi.  Is there a way to get Python to echo the contents of a script while it
> is
> being loaded?  I have a script called x.py that contains:
> 
> # x.py
> a = (1, 2, 3)
> print a
> 
> and I would like to see
> 
> a = (1, 2, 3)
> print a
> (1, 2, 3)
> 
> as the output from "python x.py".
> 

#!/usr/bin/python
"""
echo the contents of a script
"""

import sys

SCRIPT = sys.argv[0]	# no, it's something else... (find the full path of the
			# script

SCRIPT_CONTENTS = open(SCRIPT).read()

print SCRIPT_CONTENTS

# EOF

Is it also possible without reading a file?

regards,
Gerrit.



From skip at mojam.com  Tue Aug 31 10:58:33 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 31 Aug 1999 09:58:33 -0500 (CDT)
Subject: Python speed and `pcre'
In-Reply-To: <oqzoz89hrb.fsf@titan.progiciels-bpi.ca>
References: <oqzoz89hrb.fsf@titan.progiciels-bpi.ca>
Message-ID: <14283.60697.728755.190667@dolphin.mojam.com>

    Fran?ois> Hi, people.  Still learning and experimenting with Python. :-)
    Fran?ois> After having translated some code (not big, but not small)
    Fran?ois> from Perl to Python, I discover it runs ten times slower.  I
    Fran?ois> did not learn to profile yet, and I am not sure it would help,
    Fran?ois> after having read that profiling gives call counts, but no
    Fran?ois> elapsed time.  Is that right?

Fran?ois,

Can't help much with your regexp woes, but you might look at my trace module 
for timing help with your application.  Get it from 

    http://www.musi-cal.com/~skip/python/

(which may or may not work immediately due to some DNS propagation delays as 
www.musi-cal.com morphs virtual from upstate New York to California).

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From ivanlan at callware.com  Sat Aug 21 15:07:38 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Sat, 21 Aug 1999 19:07:38 GMT
Subject: Man pages would be cool.
References: <37BC6986.A66B044E@starmedia.net>
		<37BE24E0.79EA064E@montana.com>
		<37BE9EAB.CB7B1380@prescod.net>
		<37BEDBF3.BCA64E58@montana.com>
		<7pmqgq$354ai@fido.engr.sgi.com> <roy-2108991425270001@mc-as01-p02.med.nyu.edu>
Message-ID: <37BEF8FA.C510B80A@callware.com>

Hi All--

Roy Smith wrote:
> 

[bobbitt]

> 
> The only vi I know is "<esc>:q!<return>".  So far, it's been all the vi
> I've ever had to know :-)  If I ever have to work on a machine that
> doesn't have emacs, I just use ed.

_Real_ men use "cat > filename".

<and-the-reset-button-is-for-sissies,-just-pull-the-plug>-ly y'rs,
Ivan;-)
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From claird at starbase.neosoft.com  Sun Aug 22 08:29:53 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Aug 1999 12:29:53 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl>
Message-ID: <FF84952AAE191501.81D4D773B3816353.A8FE01731A222503@lp.airnews.net>

In article <7pniv5$mlm$2 at newshost.accu.uu.nl>,
Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>srenner at lycosmail.com wrote:
>> The reactions Mr. Christiansen evokes may have more to do with his style
>> than his content. It strikes me as pompous, pretentious, hostile,
>> autodidactic, preening, vain and unpleasant. Perhaps this is not the
>> right newsgroup for him.
>
>Oh, I don't think he's that bad. He *does* flame back when he feels people
>are flaming him, and sometimes his finger is a bit too near the flame
>trigger, but he's nice and reasonable to the people who are nice and 
>reasonable to him.
>
>So let's try to be nice and reasonable to him? I thought this group was
>above this kind of flaming, actually. I always considered comp.lang.python
>to be one of the most mature newsgroups I've been reading, but I'm slowly
>being forced to adjust my opinion here, and I don't like doing that.
			.
			.
			.
Maybe the newsgroup is changing.  Back to the high
ground, Pythoneers.

When did "autodidact!" become a curse?
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From g_v_wilson at my-deja.com  Tue Aug 24 22:10:17 1999
From: g_v_wilson at my-deja.com (g_v_wilson at my-deja.com)
Date: Wed, 25 Aug 1999 02:10:17 GMT
Subject: problem compiling extension on Win32
Message-ID: <7pvja1$u8e$1@nnrp1.deja.com>

I am trying to build a small Python extension
in C on Windows NT using the Microsoft 'cl'
compiler from the command line (long story).
Everything is working, except my attempts to
#define USE_DL_EXPORT and then use DL_EXPORT.
I have tried #defining the former both before
and after #including Python.h, but DL_EXPORT's
expansion is never correct.  Has anyone run
into this?  If so, I'd be grateful for help...

Please reply directly to:

 g v w i l s o n @ i n t e r l o g . c o m

as I am not a regular reader of this group.

Thanks,
Greg

p.s. anyone out there building Fortran extensions
on Windows with G77?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From sjmachin at lexicon.net  Mon Aug 16 16:51:57 1999
From: sjmachin at lexicon.net (John Machin)
Date: Mon, 16 Aug 1999 20:51:57 GMT
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
In-Reply-To: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk>
References: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk>
Message-ID: <19990816205639474.AAA315.413@max41127.izone.net.au>

On 14 Aug 99, at 19:12, Robin Becker wrote:

> I once described perl as being like awk on
> steroids and lsd

Actually perl is more like SNOBOL4 on sedatives.

There is nothing new under the sun. SNOBOL4, which was up to version 3 
by 1969, had goodies like arrays (like Python lists, but with multiple 
dimensions that were fixed when the array was created), tables (like 
Python dictionaries), user-defined functions, user-defined data-types 
(about the same functionality as C's struct), built-in pattern 
matching, operator over-riding, extensibility through "external 
functions" written in FORTRAN or assembly language, automatic garbage 
collection, ...
It even had the ability to assign the result of a partial pattern match 
to a variable ... about 30 years later, Python got this in the re 
module's ?P<name> feature, and AFAIK perl doesn't have this even yet.

However SNOBOL4 also had the ugliest syntax that I've ever seen in a 
language that was genuinely intended for serious work -- Intercal is 
far uglier but is a parody.

For example [1]
	FLIP = LEN(*I) . HEAD LEN(1) $ X LEN(1) $ Y *LGT(X,Y)
	...
	I = LT(I,LIMIT) I + 1 :S(LOOP)

That last line has this effect, expressed in C:
	if (i < limit) { i++; goto loop; }

Another "feature" was indirection.
	FOO = 'BAR'
	$FOO = 'ZOT'
gives the same result as
	BAR = ZOT
"The statement
	$INPUT = X
is an extreme example in which a string is read in and used as a 
variable" [2]

awk, which started out in 1977, has no arrays, no structs, and didn't 
get user-defined functions till 1985 [4]. Interestingly, both SNOBOL4 
and awk came from the one source: Bell Labs.

By the way, SNOBOL4's arrays had the same (documented) trap-for-young-
players as Python's:

"Each element of an array is given the same initial value. 
Consequently, execution of the instructions 
	A1 = ARRAY(5)
	A2 = ARRAY(5, A1)
creates only two arrays. Each element of A2 has the same array, A1, as 
value" . [3]

In Python, using a size of 3 instead of 5 so's it'll fit on a line:
>>> a1 = [''] * 3
>>> a2 = [a1] * 3
>>> a1, a2
(['', '', ''], [['', '', ''], ['', '', ''], ['', '', '']])
>>> a2[0][0] = 'x'
>>> a1, a2
(['x', '', ''], [['x', '', ''], ['x', '', ''], ['x', '', '']])
>>> a1[2] = 'z'
>>> a1, a2
(['x', '', 'z'], [['x', '', 'z'], ['x', '', 'z'], ['x', '', 'z']])

======================
References:
[1] Griswold, R.E., Poage, J.F., and Polonsky, I.P., "The SNOBOL4 
Programming Language", 2nd ed., Prentice-Hall, Englewood Cliffs NJ, 
1971. -- page 80.
[2] ibid, p. 192
[3] ibid, p. 114
[4] GNU gawk distribution, gawk.info, section entitled "History of 
`awk' and `gawk'"
======================

Did-anyone-we-know-write-a-SNOBOL4-mode-for-the-IBM-026-keypunch?-ly 
yours,

John






From davidopp at megsinet.net  Thu Aug 26 10:44:48 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Thu, 26 Aug 1999 10:44:48 -0400
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net>
Message-ID: <37C552DF.9F90BA36@megsinet.net>

I don't quite understand the relevance of this entire Perl advocacy string
here.  If Xah thinks that getting mean responses to postings (especially
postings from newbies) is so very cool, then why waste time over here doing
postings to the Python newsgroup.  I must say that in most newsgroups I have
found people friendly and helpful.  If I were to post to a newsgroup and get
flamed for not knowing anything or given obviously incorrect information just
to piss me off, I'd never go back.  It would not generate feelings of
advocacy for me.  Who the hell wants to have some intellectual elitist
assholes tell you you're ignorant just for asking a question?!

That being said, are there ways to integrate Python and Perl to do things
that neither one can do alone?!

David O.

Cameron Laird wrote:

> In article <37b634a2 at cs.colorado.edu>,
> Tom Christiansen  <tchrist at mox.perl.com> wrote:
>                         .
>                         .
>                         .
> >Think back to the 80s and early 90s.  The change, if it really happened,
> >came with the upty-gazillion CGI script kiddies who couldn't program
> >2+2, yet who wanted *us* to write *them* these persistent, encrypted,
> >crossplatform, multiscreen shopping carts for e-commerce, complete with
> >dynamically-generated animated vanity counters, each and every time
> >they stepped up to the feeder.
>                         .
>                         .
>                         .
> And e-mail address validation; don't forget that.
> No, wait, I mean, the *eight* things no one ever
> expects from the Spanish Inquisition are ...
> --
>
> Cameron Laird           http://starbase.neosoft.com/~claird/home.html
> claird at NeoSoft.com      +1 281 996 8546 FAX




From culliton at clark.net  Sat Aug 21 12:22:47 1999
From: culliton at clark.net (Tom Culliton)
Date: Sat, 21 Aug 1999 16:22:47 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu>
Message-ID: <rnAv3.3533$qt5.98359@iad-read.news.verio.net>

In article <37bd5e2a at cs.colorado.edu>,
Tom Christiansen  <tchrist at mox.perl.com> wrote:
>Private mail has indicated a remarkably amount of distrust about my
>learning Python.  Some verges on the unbelievable.  

Tone.  Everything has to do with tone.  If you read a comparison of
Perl to some other language written in the same tone as your original
Python/Perl messages you might bristle too.  Your tone makes it sound
like you're trashing Python, and often arbitrarily, "this isn't done
in the same way thet perl does it so it's gross..." often seems to be
the sum and total of your objection in the original message.  You do
tone it down some in the next one, and even point out things that you
think are "cool" about Python, but that write up still often misses
the point because of gaps in your understanding.

>Secondly, and much more importantly, the reason I am learning Python
>is *not* because I wish to "score some quick hits and prove to c.l.p.m
>that Perl is best after all" [sic] as one private querent asked, to my
>complete and utter horror.
>
>You may not have known this of me, but I am a collector of languages,
>and to some extent, of language trivia.  Can you honestly believe that
>I studied Spanish, French, German, Italian, Portuguese, or Latin merely
>to prove to myself and the world that English was the best language?

I would assume your motivations for learning languages are the same as
mine, to broaden your horizons, pick up useful ideas and concepts, be
able to understand what other folks are on about, ...  All noble and
admirable goals.

Your motivation, as a prominent Perl advocate, for publishing these
comparisons, while you're still coming up the learning curve, seems
more questionable, and appearance is usually all that we have to go by
on the net.

Every language has it's oddities, rough spots, and abuses.  Every
language has it's own idioms, tricks, and required knowledge.  Every
language has it's advocates and detractors.  Some of your critiques of
Python are valid, some are just incomplete knowledge, some are just
matters of taste or what you're used to.  Since you are who you are,
your comparison got widely published by LWN and possibly others as
"gospel" which is rather distressing.

Frankly I generally skip anything in this newsgroup with Perl in the
title as a likely flame fest.  I only saw your messages because LWN
had published them.  Since most of what I've seen you post in the past
seems rational, reasonable, and intelligent, my take has been that
you're somebody who has interesting and worthwhile things to say, if
with a general Perl bias.  So maybe you were just having an off day
and weren't minding your style?  ;-) ;-) ;-)

I urge you to keep trying and keep writing about your experience with
learning Python.  It's good for the Python community to get different
perspectives on the language, and discuss them.  I also urge you to
double check the details, and avoid potential flamebait when you do
so.  If we can't get by your style we can't appreciate your substance.



From kistler at fnmail.com  Sat Aug 28 07:50:30 1999
From: kistler at fnmail.com (Per Kistler)
Date: Sat, 28 Aug 1999 13:50:30 +0200
Subject: Python in Blender
References: <37C6F48C.4C465AFB@fnmail.com> <14279.22415.399945.550048@dolphin.mojam.com>
Message-ID: <37C7CD06.4DC014EF@fnmail.com>

Skip Montanaro wrote:

> No.  Do you have a URL?

http://www.blender.nl/

The new builtin Python feature is described among
the C-key features. It's a brand new feature,
which does not do that much yet. The standard modules
are not included. But one can access Blender objects
and refer to or change their geometrical properties.

An example for it's usage can be seen here:

http://www.blendermania.com/tutorials/tutor/theiceman/genodoespython.php3

The Blender newsgroups are here:

http://goethe.bowtie.nl/cgi-bin/web-ssql/news-blender/index.ws

-Per.

-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From jeremy at cnri.reston.va.us  Tue Aug 17 16:04:48 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Tue, 17 Aug 1999 20:04:48 GMT
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <14265.47748.241386.291144@buffalo.fnal.gov>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<14265.47748.241386.291144@buffalo.fnal.gov>
Message-ID: <14265.49248.386119.227023@bitdiddle.cnri.reston.va.us>

Thanks for the analysis!  I admit that metrics I used when I judged the 
randomizer I posted to be of high quality were: (1) ability to fit in
a single line and (2) obscurity.  I didn't mean to imply that the
method was good according to any other metric.

>The "hylton" method favors even permutations over odd ones, tends to
>produce increasing sequences, and distributes the variance unequally,
>apparently peaking near the beginning (1/e of the way through?).  I
>haven't quite fully analyzed the reasons for this yet...

The "shufflesort" or "Tim Peters" method, which I would offer as
preferred names (because I don't particularly want my name associated
with it and Tim's name is as good as anyone's), depends an awful lot
on the Python quicksort implementation (another good reason to call it 
the Tim Peter's method <wink>).  

I expect that the reason its tends to produce increasing sequences is
that you started with a sorted sequence; if you reverse the sequence,
it will probably tend towards decreasing sequences.  I think this
follows from the quicksort approach.  Quicksort is going to divide
each list into two sublists where the each element of the first
sublist is less than or equal to each element of the second sublist.

Since the comparison function is random, the sort is just going to
exchange a random number of elements around the pivot point.  Then
quicksort is going to be applied to each sublist, so no element from
the second sublist is ever going to be moved to the other side of the
original pivot point.  The end result is that the shufflesort method
will often leave elements close to where they were originally.

Jeremy




From mal at lemburg.com  Mon Aug 30 09:58:37 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 30 Aug 1999 15:58:37 +0200
Subject: Suitable for Large Applications?!
References: <37C7F91A.B37535@megsinet.net>
Message-ID: <37CA8E0D.3913AE16@lemburg.com>

David Oppenheimer wrote:
> 
> Dear Group,
> 
> I have just started learning Python and would like to hear more
> discussion about its suitability for large multithousand line
> apllications.  On one hand I hear people discussing applications that
> have 10,000 plus lines and on the other I hear people say its really
> only good for small tasks and for modeling.  I'd hate to spend time
> getting real good at this language only to realize that its not well
> suited to large applications...

Python's very well suited for large apps (just look at Zope for
example), it scales well, provides interfaces to many different
application domains and is by far the most readable language
I've come across in years, making it ideal for multi-developer
projects.

> Do Python programs need to be converted to C or C++ and then compiled to
> be really useful AND run fast?

No.

The normal approach is doing all the high level stuff in Python
and then possibly write extensions for low-level needs-to-be-as-fast-
as-possible parts in C/C++.

And if you don't like coding C/C++ yourself, there are quite a lot
of people around here, who would do that part for you... (plugging
my Python services here ;)

>  Or can you get away with creating a
> freestanding  executable by packaging the bytecode in freeze or another
> program like installer?

Yep.

> What are the real performance issues.  When I hear someone say that
> copiling the Python code took hours and compiling equivalent C code took
> seconds, tends to get a newbie like me less enthused.  Should I take
> back my O'Reilly Python book and get a book on learning C++?!

Compiling Python source code to Python byte code is at least as
fast as compiling and linking C code. Even better: since Python
apps are normally organized in small modules, you only need to
translate those that have changed. On to top this, Python takes
care of the process by itself -- no need to explicitly invoke
any compiler.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   123 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From philh at vision25.demon.co.uk  Wed Aug 11 17:47:18 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Wed, 11 Aug 99 21:47:18 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B02965.B1FE0E8D@prescod.net> <934326002snz@vision25.demon.co.uk> <7oraiv$nuo$1@newshost.accu.uu.nl>
Message-ID: <934408038snz@vision25.demon.co.uk>

In article <7oraiv$nuo$1 at newshost.accu.uu.nl>
           m.faassen at vet.uu.nl "Martijn Faassen" writes:
> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> [snip]
> > I am intending to do something like this. For example:
> 
> >    sub.font = "Helvetica"
> 
> > would mean: all components under the current component default to
> > using the Helvetica font.
> 
> You may want to look at Digital Creation's acquisition for this. I saw
> it mentioned once that this might be useful for GUI's too. It's used
> all over Zope and in Zope it's very powerful.

I really will have to learn Zope one of these days.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From aaron_watters at my-deja.com  Fri Aug 20 09:26:49 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Fri, 20 Aug 1999 13:26:49 GMT
Subject: sliciing dicts (was RE: More random python observations from a perl programmer)
References: <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <001301beeacf$9819a560$712d2399@tim>
Message-ID: <7pjl2c$npl$1@nnrp1.deja.com>

In article <001301beeacf$9819a560$712d2399 at tim>,
  "Tim Peters" <tim_one at email.msn.com> wrote:
> [tchrist]
> >> GOTCHA: (medium)
> >>     Slices in Python must be contiguous ranges of a sequence.
> >>     In Perl, there's no such restriction.
>
> [Michael Hudson]
> > At some point Python will probably support `extended slice syntax'
> > like so:
> >
> > range(0,10)[1:4:2] => [1,3]
> >
> > Numerical Python already does this.
>
> [tchrist]
> >> GOTCHA: (medium)
> >>     You can't slice dictionaries at all in Python.  In Perl, it's
> >>     easy to slice a hash, and just as sensible.
>
> [Michael Hudson]
> > Huh? What would that mean? There's no reason to suspect that the
keys
> > of a dictionary are ordered.
>
> What Tom calls "slicing" here the NumPy folk would call gather/scatter
> subscripting.  A list can be used as an index, e.g.
>
>     thing[[7,2,3]] == [thing[7], thing[2], thing[3]]
>
> Perl has something like that; Python does not.

All of these things of course can be implemented using the
various python meta-hooks either at the Python or the C level.
They aren't intrinsic limitations of the Python interpreter,
just limitations of the standard dictionary/list/tuple
implementations.

BTW, kjbuckets.kjDict's have a method called "dump" (with a dual
kjUndump function) where

    D.dump((x,)) --> D[x]
    D.dump((x,y,z)) --> (D[x], D[y], D[z])

Gadfly uses these features to quickly pack/unpack data in/out
of dictionary format either for indexing or archiving.

I really think you should get kjbuckets, Tom, it sounds like it's just
what you need for the Python app you are apparently working on. :)

   Aaron Watters http://www.chordate.com

===

I forgot my mantra.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From wware-nospam at world.std.com  Tue Aug  3 10:31:08 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Tue, 3 Aug 1999 14:31:08 GMT
Subject: Python microthreads
References: <FFropB.7tp@world.std.com> <933672146.479670@news.adtranzsig.de>
Message-ID: <FFw8Bw.CI4@world.std.com>

Dirk-Ulrich Heise (hei at adtranzsig.de) wrote:
: Did you build it on top of Christian Tismers
: Stackless Python, or is it a different kind of
: modification to the VM?

I haven't looked at Stackless Python, but it sounds like an unrelated
thing. As I recall, Stackless Python uses the C stack for everything
whereas normally Python has a piece of stack in each frame object.
It can do this because the compiler is smart enough to figure out
exactly how much stack will be needed for any function call. So
frameobjects have a field called f_valuestack which maintains the
arguments and local variables for that function.

What I did was to keep the frameobject exactly as it is in normal
Python, but invent a thread object with a stack for frameobjects,
and a couple more stacks for stack pointers and program counters.
Additionally, I modified the bytecode interpreter (eval_code2, in
ceval.c) so that it runs for a specified number of opcodes and
then stops, leaving the entire environment intact. This lets you
rotate thru several threads, executing each for a few opcodes, so
it looks like the threads are running simultaneously.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From fredrik at pythonware.com  Thu Aug 26 04:47:45 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 26 Aug 1999 08:47:45 GMT
Subject: idling with tkinter
References: <7q2u6e$aso$1@nnrp1.deja.com>
Message-ID: <004901beef9f$ac5da790$f29b12c2@secret.pythonware.com>

G. Lewis <zontar4 at yahoo.com> wrote:
> I found the after_idle() method, but it only calls the callback
> once. Strangely i can't find a method to perminently hook a
> callback into the idle system. Is there one?

yes.  the appropriate method is to call the 'after'
method from inside the callback.  to get things
going, just call the callback once.

you'll find sample code at:
http://www.pythonware.com/library/tkinter/introduction/basic-widget-methods.htm
(click on "alarm handlers and other non-event callbacks")

</F>




From graham at sloth.math.uga.edu  Mon Aug 30 10:13:25 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 30 Aug 1999 14:13:25 GMT
Subject: Suitable for Large Applications?!
References: <37C7F91A.B37535@megsinet.net>
Message-ID: <7qe3i5$kok$1@cronkite.cc.uga.edu>

David Oppenheimer (davidopp at megsinet.net) wrote:
: I have just started learning Python and would like to hear more
: discussion about its suitability for large multithousand line
: apllications.  On one hand I hear people discussing applications that
: have 10,000 plus lines and on the other I hear people say its really
: only good for small tasks and for modeling.  I'd hate to spend time
: getting real good at this language only to realize that its not well
: suited to large applications...
: Do Python programs need to be converted to C or C++ and then compiled to
: be really useful AND run fast?  Or can you get away with creating a
: freestanding  executable by packaging the bytecode in freeze or another
: program like installer?
: What are the real performance issues.

I think you are slightly confused. Whether or not Python is well
suited to large projects has very little to do with performance
issues. Indeed this is true for any language. Large projects are
dominated by software engineering issues, and not performance issues.
I personally don't find Python good for large projects because of
its lack of type checking, and its lack of garbage collection (this
is mitigated to some extent by it's extensive library). Your mileage
may vary though.

graham
-- 

      England's not the mythical land of madame george and roses
        It's the home of police who kill black boys on mopeds



From da at ski.org  Mon Aug 16 15:50:01 1999
From: da at ski.org (David Ascher)
Date: 16 Aug 1999 14:50:01 -0500
Subject: README: Python-URL! - the weekly guide to Python resources (Aug 16)
Message-ID: <C7374D44CCACDC61.D9C6108323B94F88.79E977ACD97F8278@lp.airnews.net>

Python-URL is back!  

Here are the latest news from the Python world.

>From the hacks & ports category:

 * Will Ware dove into the Python core to allow microthreads (threads
   of operation which work in a single system thread).  This is
   similar in spirit to some of work done by Christian Tismer for his
   "Stackless Python", and there are rumors that the former will be
   adapted to the latter.

	http://world.std.com/~wware/uthread.html
	http://www.pns.cc/stackless/stackless.htm    
	http://www.deja.com/getdoc.xp?AN=507505476

 * Python is routinely ported to embedded systems.  A Finnish group is
   making their port public, and soliciting feedback on this work:

	http://www.abo.fi/~iporres/python/

 * Greg Ewing puts forth a patch to add lexical scoping to Python.
   
        http://x36.deja.com/viewthread.xp?AN=510811241
        http://www.cosc.canterbury.ac.nz/~greg/python/lexscope.html

 * Max Skaller apparently has a new implementation of Python in the
   works with speed in mind:

        http://x42.deja.com/[ST_rn=md]/viewthread.xp?thitnum=3&mhitnum=0&toffset=0&CONTEXT=934822861.1754792007&frpage=threadmsg_md.xp&back=comp.lang.python&new=0&rok=1

>From the gotta-have-a-usenet-discussion-every-now-and-then-discussion:

 * for + else = ???

        http://www.deja.com/viewthread.xp?AN=505752853

 * YAPPW (Yet Another Perl Python War) starts at:

        http://x42.deja.com/[ST_rn=md]/threadmsg_md.xp?thitnum=1&mhitnum=1&CONTEXT=934822861.1754792007

        http://x42.deja.com/[ST_rn=md]/viewthread.xp?thitnum=0&mhitnum=0&toffset=0&CONTEXT=934822861.1754792007&frpage=threadmsg_md.xp&back=comp.lang.python&new=0&rok=1

 * A new keyword is almost proposed, from of all people Tom Christiansen:

	 http://x42.deja.com/[ST_rn=md]/viewthread.xp?thitnum=24&mhitnum=0&toffset=0&CONTEXT=934822861.1754792007&frpage=threadmsg_md.xp&back=comp.lang.python&new=0&rok=1

>From the power tools department:

 * The latest SWIG release has support for Python keyword arguments
   and better shadow class support:

       http://www.egroups.com/group/swig/1181.html?

 * Python Imaging Library (PIL) version 1.0 is out:

       http://www.pythonware.com/products/pil/

 * Too many news in the Zope front to go into in detail, but foremost
   is a new site layout and several Zope 2.0 beta releases:

       http://www.zope.org/

 * Numeric Python release 12 beta now uses ExtensionClasses:

       ftp://ftp-icf.llnl.gov/pub/python/LLNLDistribution12beta.tgz

 * Paul Dubois' PyFort is a tool for interfacing NumPy to Fortran:

       http://www.deja.com/getdoc.xp?AN=510814523

 * SCXX is a lightweight take on Paul Dubois's CXX by Gordon McMillan
   which promises to require less of your C++ compiler:

       http://www.deja.com/getdoc.xp?AN=510361661

>From the events and public relations department:

 * Local Python users groups are sprouting, and they seem to be called
   PIGgies!.  The San Francisco Bay Area has one, as does the
   Washington DC area.  Where will the next one crop up?

       http://www.baypiggies.org/
       http://www.python.org/Events.html#dcpiggies

 * Bruce Eckel, a C++ and Java trainer and book author, has picked up
   the Python bug, and is planning what sounds like a neat Python
   workshop -- learn Python in five days in mountains by writing
   something useful.

	http://www.eckelobjects.com/PythonProjectWorkshop/index.html

 * The O'Reilly conference is soon upon us:

	http://conferences.oreilly.com/python/   

 * The next International Python Conference has a revised call for
   papers with new dates:

	http://www.python.org/workshops/2000-01/cfp.html


=========================================================================
Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

    Archive probing trick of the trade:
        http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
  http://www.egroups.com/list/python-url-leads/

To receive a new issue of this posting in e-mail each Monday morning
(once we resume), ask <claird at neosoft.com> to subscribe.


-- The Python-URL! Team--
-David Ascher <da at ski.org>
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From eugene.leitl at lrz.uni-muenchen.de  Wed Aug 11 13:21:13 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Wed, 11 Aug 1999 10:21:13 -0700 (PDT)
Subject: converting Netscrape mailbox to Unix mbox format
In-Reply-To: <m3wvv24wcg.fsf@atrus.jesus.cam.ac.uk>
References: <14256.35316.328575.118034@lrz.de>
	<m3wvv24wcg.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <14257.45321.128760.603105@lrz.de>

Michael Hudson writes:
 
 > Last I looked Netscape mailbox format *was* unix mbox format (on unix
 > anyway; can't speak for Netscape on windows) - have you tried pointing
 > pine at them?
 
Yes. Both pine and VM. Both see a giant 5 MByte message.
 
 > Otherwise you could try the mailbox standard library module...



From nickb at earth.ox.ac.uk  Mon Aug  2 05:51:20 1999
From: nickb at earth.ox.ac.uk (Nick Belshaw)
Date: Mon, 02 Aug 1999 10:51:20 +0100
Subject: ioctl and mmap question
Message-ID: <37A56A18.A54119FA@earth.ox.ac.uk>

I would be very grateful if anyone direct me on an issue which  is
confusing me a little?

I am playing with a Linux device driver which uses DMA on a PC using
RedHat Linux.

Installation of the module is fine but when I open the device from
Python I don't get all I want. I have ioctl access which seems to allow
all the code suggests and read/write sort of work. The module has a
mmap-fop also and I don't see how or when to call it. Is it intended to
be called automatically when I do other things or do I have to deal with
it myself? If I have to call it explicitly how do I do that from Python?
I have not seen any references to mmap calls to devices and am slightly
confused.

thanks in advance

Nick/Oxford





From cbbrowne at news.hex.net  Thu Aug 19 20:52:09 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Fri, 20 Aug 1999 00:52:09 GMT
Subject: Unsorting(randomizing) a sequence
References: <000701bee917$d8161cc0$eaa22299@tim> <14268.10663.137315.893101@buffalo.fnal.gov>
Message-ID: <ZE1v3.14955$gO1.492954@news2.giganews.com>

Is it so much work to come up with a half-reasonable card shuffling
algorithm?

The one I've always seen cited as providing decent behaviour looks
thus:

for (i = 0; i > n; i++) {
   swap (item[i], item[random(n)]);
}

This algorithm guarantees that each and every item on the list gets
swapped with something.

If I wanted to make sure the permutation *really* looks random, I'd
consider running this "shuffle" multiple times, and would make sure I
had a RNG of decent quality with a period Rather Longer than n.
-- 
"For systems, the analogue of a face-lift is to add to the control
graph an edge that creates a cycle, not just an additional node."
-- Alan Perlis
cbbrowne at hex.net- <http://www.hex.net/~cbbrowne/languages.html>



From Ted.Horst at wdr.com  Thu Aug 26 17:03:54 1999
From: Ted.Horst at wdr.com (Ted Horst)
Date: Thu, 26 Aug 99 16:03:54 -0500
Subject: fully qualified hostname ?
In-Reply-To: <595C6D6DD3292BB3.DCF1BAEEAE3E421A.EFDF99C4B418A66D@lp.airnews.net>
References: <199908261822.AA01099@ch1d2833nwk>
	<595C6D6DD3292BB3.DCF1BAEEAE3E421A.EFDF99C4B418A66D@lp.airnews.net>
Message-ID: <199908262103.AA01157@ch1d2833nwk>

Thanks for the rapid response!

On 26 Aug 1999, claird at starbase.neosoft.com (Cameron Laird) wrote:
> In article <199908261822.AA01099 at ch1d2833nwk>,
> Ted Horst  <Ted.Horst at wdr.com> wrote:
> >
> >What is the best way to get the fully qualified hostname in python (ie
> >hostname.domainname) ?
> >
> >A portable solution would be ideal, but I would settle for Unix only.
> .
> .
> .
> There's a lot of ambiguity in the question of which
> you're not aware.  A perfectly well-behaved network
> host might have several fully-qualified hostnames,
> or none.  One can even make a case that a single
> host is best known by different names on different
> ports.
>

OK, I can see that.  If there are multiple valid choices (meaning that they  
would resolve ok outside of my domain), then I don't think I care which one  
I get.

> So:  you're definitely entering the land of some-
> what sloppy heuristics.  The first thought that
> occurs to me is a reverse DNS lookup; is that ade-
> quate to your situation?

That sounds reasonable to me, but this is not really my area of expertise.   
How would I do this from python ?

>
> Note, also, that many Python instances run on host-
> nameless hosts.

The context here is distributed objects, CORBA for now.  So how would you  
encode the host for inclusion in an IOR if there is no hostname ?

> --
>
> Cameron Laird           <http://starbase.neosoft.com/~claird/home.html
> claird at NeoSoft.com      +1 281 996 8546 FAX
>
> --
> <http://www.python.org/mailman/listinfo/python-list
---



From arcege at shore.net  Wed Aug 11 12:14:10 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 11 Aug 1999 16:14:10 GMT
Subject: Why is CGI module so complicated?
References: <37AB04A8.C3403427@NOSPAM.strs.co.uk> <cWYr3.1329$EG4.226668@news.shore.net> <37B15BF8.B7204705@strs.co.uk>
Message-ID: <mjhs3.1462$EG4.260559@news.shore.net>

Ian Clarke <I.Clarke at strs.co.uk> wrote:

:> : Can anyone explain what the point of the FieldStorage and
:> : MiniFieldStorage objects are?

:> Ian, please read my reply in the previous thread:
:>   http://www.dejanews.com/getdoc.xp/AN=502365247

: Well, unfortunately that link didn't work, but I managed to find the
: comment I believe you are referring to, in which case I assume you were
: indicating that the FieldStorage and MiniFieldStorage objects were used
: for the case where a file upload is taking place.

Nasty, dejanews changed the article number on me (I tested this with
lynx before I posted my article).  But yes, that is the article.

The _exact_ URL that I have this morning (8-11-99 11:52EDT) is:
http://x21.deja.com/getdoc.xp?AN=511078528&CONTEXT=934386589.945881145&hitnum=0

: I can't help but feel that this is not the best way.  How about go with
: the method I described in my first post (where cgi fields are returned
: as the types that they actually are, rather than being encased in
: MiniFieldStorage objects).  Were more than one item is selected for a
: given field a list could be returned, and for a file upload a tuple
: could be returned containing the filename and the contents of the file -
: or if that is not considered extendable enough, a CGIFileUpload object
: could be returned containing the necessary fields.

I wondered why myself at first too, especially since a list is returned
instead of an object that looks like a list (with a value attribute),
when there are multiple fields returned (checkbuttons for example).

I usually have my own routine that converts the objects to a UserDict
(or a real dict) after I run them, unless I expect a file-upload in the
form.  It's nothing major, but it helps.  It's better to leave the
interface as it is (all fields return the same interface),  I think it
works a little better in a HTML 4.0 view of the world.

: This would mean that for the vast majority of CGI fields which are just
: normal text, retrieval is very straight-forward and intuitive, just like
: accessing strings stored in a dictionary.  More complexity is only
: introduced when the user wants to do more complicated stuff (and even
: then, it is pretty simple).  As mentioned in my previous post, it would
: also allow the fields to be used as the substitutions for a
: TemplateDocument directly, which can be quite a useful trick.

There is a lot of information kept in the FieldStorage object which may
be desirable (like Content-Type parameters, Content-Length, etc.).
When using multipart/mixed instead of x-www-form-urlencoded, each field
(including non file-upload fields) can have these attributes.  The
existing interface does not require a _lot_ of special coded to handle
the different cases in the CGI protocol.

: If I get some spare time today I will create a small module which will
: provide this interface to CGI (although if anyone else wants to do it,
: feel free!).  It would obviously be better if (assuming I can convince
: others that this is a better way) this was integrated into the built-in
: cgi module.

I think the existing interface is fine, but a "simple" translater
around things might be good; something like:

  def FieldStorage2Dict(form):
    d = {}
    def decode_field(field):
      if isinstance(field, type([])):
        return map(decode_field, field)
      elif hasattr(field, "file") and field.file:
        return (field.filename, field.file)
      else:
        return field.value
    for key in form.keys():
      d[key] = decode_field(form[key])
    return d

What would be _really_ good (IMO) is to have a "toDict" method that
does this.

  -Arcege




From moshez at server.python.net  Tue Aug 31 01:27:35 1999
From: moshez at server.python.net (Moshe Zadka)
Date: Tue, 31 Aug 1999 01:27:35 -0400 (EDT)
Subject: 'if foo' calls getattr?!?
In-Reply-To: <1276085477-52750363@hypernet.com>
Message-ID: <Pine.LNX.4.10.9908310124220.9802-100000@starship.python.net>

On Mon, 30 Aug 1999, Gordon McMillan wrote:

[about how hard it is to know the truth in Python]
> In which case "if foo is not None" will be the fastest and 
> closest-to-your-intention test possible. (But "if foo != None" will 
> get you right back to where you are now!).

Bzzt! foo!=None is slower then foo is not None, but not by much: it will
first check whether type(foo) is not type(None), so it is pretty fast.

truth-and-nothing-are-always-popular-subjects-ly y'rs, Z.




From arcege at shore.net  Tue Aug 17 11:27:45 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 17 Aug 1999 15:27:45 GMT
Subject: Getting the module name from a method...
References: <7pboln$n74$1@feed.teaser.fr>
Message-ID: <Rbfu3.385$mV.141087@news.shore.net>

Olivier Deckmyn <olivier.deckmyn at mail.dotcom.fr> wrote:
: Hi !

: I would like a method of my own to be able to print the module it is
: declared in:

: ex:
: the text for the module : toto.py is :

: class MyClass:
:     def myMethod(self):
:         print "I am in module", ????????

: and then
: MyClass().myMethod()
: should produce :
: 'toto'

Unless the method was added as an instance attribute (self.method = ...),
the global namespace should be the module where the method was defined,
in this case "toto".  The less secure means would be
globals()['__name__'], but you could also go with:

  import os, sys
  def mymodule():
    try:
      raise RuntimeError
    except RuntimeError:
      exc, val, tb = sys.exc_info()
      code = tb.tb_frame.f_back.f_code
      del exc, val, tb
    if code.co_filename:
      module = os.path.splitext(os.path.basename(code.co_filename))[0]
    else:
      module = '<string>'
    return module

  -Arcege




From frehe491 at student.liu.se  Sat Aug 21 09:11:27 1999
From: frehe491 at student.liu.se (Fredrik Henbjork)
Date: Sat, 21 Aug 1999 15:11:27 +0200
Subject: What is a troll by the way? [offtopic]
References: <Z1xv3.11$EN1.7779@news.shore.net>
Message-ID: <99082115204502.12287@o112>

On Sat, 21 Aug 1999, you wrote:
> Chad Netzer <chad at vision.arc.nasa.gov> wrote:
> : Robert Kern wrote:
> 
> :> _Troll_ is a USENET term for someone that deliberately posts an
> :> inflammatory message simply to stir up argument.
> 
> : I've found it funny to watch this term evolve over the past few years.
> : It started (I believe) as a reference to the fishing term, and used to
> : mean "trolling for flames" (ie. hoping to bring flames to the
> : discussion), perhaps employing "flame bait" for this purpose.
> 
> Anyone remember BIFF?

WH0 KANN F0R63T ZUCH AN AWZUM K001 D00D!!!!! BIFF R00LZ!!!!
If you miss him, just head over to the discussions at Slashdot. The only
difference is that the ones at Slashdot are *real* <shudder>.

--
Fredrik Henbjork

Email: frehe491 at student.liu.se
WWW: http://o112.ryd.student.liu.se



From fredrik at pythonware.com  Tue Aug 31 06:06:39 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 31 Aug 1999 10:06:39 GMT
Subject: Bug in Tkinter?
References: <c75zoz849tq.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <010401bef398$85ae3e00$f29b12c2@secret.pythonware.com>

<tre17 at student.canterbury.ac.nz> wrote:
> Have a look at the following code.  It should produce a window
> containing two radio buttons, one of which is active when the program
> starts.  But it does not!  Now remove the comments, so that there is a 
> permanent reference to StringVar instance, and now it works?

feature, not bug.  it's related to:
http://www.python.org/doc/FAQ.html#4.69

</F>





From grant at nowhere.  Wed Aug 25 15:03:50 1999
From: grant at nowhere. (Grant Edwards)
Date: Wed, 25 Aug 1999 19:03:50 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7pp517$dce$1@nnrp1.deja.com>
Message-ID: <slrn7s8fgl.d6.grant@grante.comtrol.com>

In article <7pp517$dce$1 at nnrp1.deja.com>, Georg Mischler wrote:
>philh at vision25.demon.co.uk wrote:
>[...]
>> To use Parrot, first you create a *.par file describing your
>> application's GUI. Your file might look something like this:
>>
>> -----------Cut here-------------------------------
>> window @MyWindow "My First Window" {
>>    menuBar {
>>       menu "File" {
>>          menuItem @New "New"
>>          menuItem @Exit "Exit"
>>       }
>>       menu "Help" {
>>          menuItem @About "About..."
>[...]
>
>
>Looks very interesting.
>And actually reminds me a lot of autodesks slowly deprecated
>dcl description language which I also liked a lot at the time.
>Is this a coincidence?

The text-based specification langauge methodology is also used
by the GUI package for Modula-3.  The interface is defined in a
text file by building up composits of controls, containers
(vertical and horizontal) and glue (hbox,vbox,glue: very
TeX-like -- with Lisp s-expressions as the syntax).  The result
can be parsed at compile-time and stored in the executable, or
parsed at run-time.  Or, if you're feeling particularly
masochistic, you can generate the s-expressions on-the-fly in
your program and parse them as you go.

It also has a prototyping application that lets you
interactively edit the interface specification and try it out.
Though I've only used it under Unix/X11, the UI Source files
can be used without change under the Windows-NT implementation
of M3.

A very well though-out system, IMHO.

-- 
Grant Edwards                   grante             Yow!  How do I get HOME?
                                  at               
                               visi.com            



From robin at jessikat.demon.co.uk  Sat Aug 21 11:25:41 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 21 Aug 1999 16:25:41 +0100
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com>
Message-ID: <jxVZvEA1Tsv3Ew7p@jessikat.demon.co.uk>

In article <7pmdn5$lt0$1 at nnrp1.deja.com>, srenner at lycosmail.com writes
>The reactions Mr. Christiansen evokes may have more to do with his style
>than his content. It strikes me as pompous, pretentious, hostile,
>autodidactic, preening, vain and unpleasant. Perhaps this is not the
>right newsgroup for him.
>
>sr
>
>
>Sent via Deja.com http://www.deja.com/
>Share what you know. Learn what you don't.
gor bli'me guv and I thought I was vicious :)
-- 
Robin Becker



From achau99 at yahoo.com  Fri Aug 20 13:51:07 1999
From: achau99 at yahoo.com (Alan Chau)
Date: Fri, 20 Aug 1999 17:51:07 GMT
Subject: subscribe
Message-ID: <19990820175107.19758.rocketmail@send205.yahoomail.com>


__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com




From robin at alldunn.com  Wed Aug 18 21:48:45 1999
From: robin at alldunn.com (Robin Dunn)
Date: Wed, 18 Aug 1999 18:48:45 -0700
Subject: Python Sighting
References: <slrn7rlrsu.l2e.mryan@unix2.netaxs.com>
Message-ID: <boJu3.107$zk2.15925592@news.randori.com>

> I made an interesting discovery the other day while roaming through the
> registry of an NT system trying to troubleshoot a problem in a Microsoft
> Site Server class I was supporting:  Microsoft Site Server 3.0 (not sure
> if it's the normal or commerce edition) uses Python.  From the presence of
> the ni.pyc file, I'm going to guess that it's 1.4.x.  I'm also not sure of
> what capacity they're using it in.
>

My understanding is that when MS asimilated the company that originally made
Site Server (I think it was Commerce Server then) that it was written in
mostly Python and stayed that way for at least the first release under the
MS name.

It's too bad they didn't leave it in Python, but at least they didn't try to
dominate and then break the Python world like they have tried to do with
nearly everything else they have touched.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!






From chinaski_py at my-deja.com  Sun Aug 22 05:14:09 1999
From: chinaski_py at my-deja.com (chinaski_py at my-deja.com)
Date: Sun, 22 Aug 1999 09:14:09 GMT
Subject: Need help with Tkinter
Message-ID: <7pof0u$av$1@nnrp1.deja.com>

When I try to import Tkinter from PythonWin this is the message I get:
>>> from Tkinter import
>>> w = Button(text = "hello", command ='exit')

Traceback (innermost last):
  File "<interactive input>", line 1, in ?

 File "C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 1123, in
__init__
    Widget.__init__(self, master, 'button', cnf, kw)
  File
"C:\Program Files\Python\Lib\lib-tk\Tkinter.py", line 1078, in __init__

   BaseWidget._setup(self, master, cnf)
  File "C:\Program
Files\Python\Lib\lib-tk\Tkinter.py", line 1055, in _setup

_default_root = Tk()
  File "C:\Program
Files\Python\Lib\lib-tk\Tkinter.py", line 886, in __init__
    self.tk =
_tkinter.create(screenName, baseName, className)
TclError: Can't find a
usable init.tcl in the following directories:
    {} ./lib/tcl8.0
C:/tcl8.0/library {C:/Program Files/library}



This probably means that
Tcl wasn't installed properly.

Can anyone explain how to fix this?
   Your help is appreciated,
                G.Norton (chinaski)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From duncan at rcp.co.uk  Thu Aug 19 12:37:50 1999
From: duncan at rcp.co.uk (Duncan Booth)
Date: 19 Aug 1999 16:37:50 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <8E27B3FC4duncanrcpcouk@news.rmplc.co.uk>

tchrist at mox.perl.com (Tom Christiansen) wrote in 
<37bc1339 at cs.colorado.edu>:
>GOTCHA: (medium)
>    The whole mutability versus immutability thing causes many 
>    confusions.
Yup, powerful but confusing at first.


>GOTCHA: (medium)
>    If you have a function that expects two arguments, such as:
>     def fn(x,y):
>         return x + y
>    and you have a tuple of the two elements:
>     t = (1,2)
>    You can't just call that function:
>     print fn(t)
>    Or you get a "TypeError: not enough arguments; expected 2, got 1"
>    error.  You have to use this apply() function to get around this
>    issue:
>     print apply(fn,t)
>     3
Or if you know you are going to have a tuple, you could declare the 
function to take a tuple as its argument:
def fn((x,y)): ...

>GOTCHA: (low) 
>    There are no compound assignment operators, like +=.
We would have to reach agreement first on what += actually meant, and that 
doesn't seem to be happening for all the number of times it has been 
discussed.


>GOTCHA: (low)
>    Single and double quoted strings can't cross line boundaries.
>    You need triple quotes for that!
I would put this one under COOLNESS.

>DISSIMILARITY:
>    Because there are no variable markers in Python, there are no 
>    variable interpolation in strings.  That makes something like
>     print "this $x is $y here\n" 
>    become in python
>     print "this %d is %d here\n" % (x, y)

You didn't mention the alternative way of writing this which is possibly 
nearer the Perl version:
print "this %(x)s is %(y)s here\n" % {'x': x, 'y': y}

>GOTCHA: (high)
>    Python has no manpages!  The horror!!!!!!!!!!!!!!!!!!!!!!
>    ENODOC
Python has loads and loads of documentation.

>GOTCHA: (medium)
>    All ranges are up to but *not including* that point.  So range(3)
>    is the list [0,1,2].  This is also true in slices, which is the
>    real gotcha part.  A slide t[2:5] does not include t[5] in it.
I think this is necessary to allow all of the variations on slicing. For 
example to insert between elements of a list.
    	x = [1, 2, 3]
    	x[1:1] = [5]
    	print x
    	[1, 5, 2, 3]

>GOTCHA: (high)
>    This is a big surprise: strings are not atomic (although they are
>    immutable).
>     >>> for c in ("weird", "bits"):

>     >>> for c in ("weird"):
For consistency with the first example you need
>>> for c in ("weird",)

However I agree it can be confusing that when you subscript a string you 
get another object of type string.

>GOTCHA: (high)
>    As we saw with lists, because everything is a reference, and there's
>    no way to dereference that reference, this means that again there
>    is also no built-in, intuitive way to copy a dictionary.  Instead, 
>    the suggested work-around is to write a loop:
>     new = {}
>     for key in old.keys:
>         new[key] = old[key]
My suggested way to copy a dictionary is:
    	new = old.copy()
Or, if you want to copy all the elements as well:
    	from copy import deepcopy
    	new = deepcopy(old)

>    But this is guaranteed slower, because it's not at the C level.
The copy method and functions are at the C level.

>    It also shows that dictionaries aren't first class citizens in python
>    as they are in Perl:
I think there is a fundamental misunderstanding here. Variables in python 
are always references. They never copy the value they point to when 
assigned. You could equally say that this means that integers are not first 
class citizens because:
    	x = 1
    	y = x
leaves x and y pointing at the same object (x is y) is true.
    	
>
>GOTCHA: (high)
>    Lists don't autoallocate the way dictionaries do.  So while this works
>    on two dictionaries:
>     new = {}
>     for key in old.keys:
>         new[key] = old[key]
>    This fails on two lists:
>     new = []
>     for i in old:
>         new[i] = old[i]
This is meaningless and nothing to do with autoallocating. i is iterating 
over values not indexes. you get the effect you want here by writing
    	for i in old:
    	    	new.append(i)

>GOTCHA: (medium)
>    Importing a variable only gives you a local copy.  In Perl, it makes
>    two names for the same object.
No, importing just gives you two names for the same object just as 
assignment only ever gives you two names for the same object. Of course if 
you later reassign one of the names to point to a different object the two 
variables no longer have the same value.

>GOTCHA: (high)
>    Because you can't use readline() to get a number, people seem to enjoy
>    calling eval() just to get a string turned into a number:
>     import sys
>     str = sys.stdin.readline()
>     num = eval(x)
>    This is scary.  Even scarier is the propensity for calling input(),
>    which auto-eval()s its input to make sure it's the right "type".
>    (Funny that a soi-disant "typeless language" should be so danged
>    picky about this.)  That means you see people write:
>     num = input("Pick a number? ")
>    But the person can supply as an answer 0x23 or 2+9 or pow(2,4)
>    or os.system("rm -rf *").  I am stunned.

People may enjoy this but personally I would just do
    	num = int(sys.stdin.readline())
BTW, I am not commenting on the comparisons with Perl since I don't claim 
to know that language, but I seem to remember hearing that Perl would 
silently convert the string "42x" into the number 42 without throwing an 
error, now there is scary.

>GOTCHA: (low)
>    Regexes default to emacs style, not egrep style!  Gads!  This is
>    surprising to anyone except an emacs weenie.  Sigh.  And just *try*
>    to read the manpage on the differences based on passed in arguments
>    establishing the style.  There aren't any manpages!  Have a nice day.
Try reading the manual page doc/lib/re-syntax.html in your python 
distribution.


>GOTCHA: (low)
>    Normally, a print statement in python adds a newline.  If you
>    don't want one, you need a trailing comma!
Or use the write method of the file which doesn't do any of the print 
prettifying.

>GOTCHA: (medium)
>    Anything that python doesn't like, it raises an exception about.
>    There is no fail-soft.  Even non-exception issues raise exceptions.
>    It's pervasive.  K&P curse languages that force people who want
>    to open() a file to wrap everything in exception code, saying that
>    "failing to open a file is hardly exceptional".  
I love this. But thats just me.

>
>GOTCHA: (medium)
>    You can't just interchange tuples and lists in python the way
>    you can lists and arrays in Perl.
Or the way you can interchange strings and numbers in Perl.
They are different types. (Why we need two list types is another question 
though).

>GOTCHA: (low)
>    You have to compiled the definition for a  function before you may
>    compile the call it.
Not true, you have to compile the definition for a function before you may 
execute the call to it. You may compile the call before the function if you 
wish.

>    ADDENDA: It turns out that def() happens *AT RUN TIME* in python.
>    That's the root of this problem.  In Perl, a sub{} is a compile-time
>    thing.
Dont forget that in Python you can stick your def in a for loop to define a 
whole host of functions with different default arguments.

Thanks for the article though, it made good reading.
(BTW if anyone flames me for getting bits of this wrong don't expect a 
reply soon because I'm away for two weeks).




From MHammond at skippinet.com.au  Thu Aug  5 18:45:22 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Fri, 6 Aug 1999 08:45:22 +1000
Subject: ADO example code failed:hlp!
References: <37A9F366.EA01D915@bigfoot.com>
Message-ID: <7od46d$h02$1@m2.c2.telstra-mm.net.au>

Hoon Yoon wrote in message <37A9F366.EA01D915 at bigfoot.com>...
>Running Christian's code, I get following error message. What am I doing
>wrong? It's same code, but the constants do not contain adModeReadWrite.
>Pls help!
>Thanks much in Adv,

Most likely you have not run "makepy" over the ADO library.

Try:

>import win32com.client
>db=win32com.client.Dispatch("ADODB.Connection")

print db

If it says "COM object <ADODB.Connection>", you have not used makepy.  If it
says "win32com.gen_py.blah..." then it is a different problem.

To run makepy, select it from the Pythonwin tools menu, or double-click on
"Python\win32com\client\makepy.py"

Mark.






From spamfranke at bigfoot.de  Mon Aug 30 07:03:57 1999
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Mon, 30 Aug 1999 11:03:57 GMT
Subject: How to change default behaviour of a class?
References: <37CA5E90.28D60E87@xsoft.co.at>
Message-ID: <37ca620d.2193924@news.omnilink.de>

On Mon, 30 Aug 1999 12:36:00 +0200, "Gerhard W. Gruber" <g.gruber at xsoft.co.at> wrote:

>How can I change what is printed when I use the statement "print class"?
>Usually this leads to an output like <class instance at 9aad60> but I
>rather want it to print variables contained in my class instead of the
>pointer adress.

>From your output I suppose meant "print instance" and "variables contained 
in my instance".

There are two special methods __repr__ and __str__, which can be used
to customize the classes string pepresentation. The print statement 
tries to call __str__. If __str__ is not defined it uses __repr__ or
the default representation otherwise.

See http://www.python.org/doc/current/ref/customization.html#l2h-128
for some documentation. 

A short example:

>>> class C:
...   def __repr__(self):
...     return repr (self.__dict__)
... 
>>> c = C()
>>> c.a = 1
>>> c.b = "My String"
>>> c
{'b': 'My String', 'a': 1}


Stefan




From skip at mojam.com  Mon Aug 30 09:47:58 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 30 Aug 1999 08:47:58 -0500 (CDT)
Subject: The (un)suitability of Make for Python projects (was: Re: Returning none)
In-Reply-To: <37C9F196.246EE7A7@prescod.net>
References: <37C55044.8CA02530@prescod.net>
	<37C5683D.91D61FD0@rubic.com>
	<14277.28442.44107.591412@dolphin.mojam.com>
	<37C70821.92E810FC@prescod.net>
	<14279.9024.401900.142461@dolphin.mojam.com>
	<37C7B86E.647BA654@prescod.net>
	<14281.61572.191617.870559@dolphin.mojam.com>
	<37C9F196.246EE7A7@prescod.net>
Message-ID: <14282.35531.241143.97477@dolphin.mojam.com>

    Paul> I know all about make, but in an all-Python project (the best
    Paul> kind) it is not much value.

I respectfully disagree.  There is much more to most successful projects
than building executables.  Building end-user documentation, running tests,
generating TAGS files, etc. are all appropriate in an all-Python project.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From pointal at lure.u-psud.fr  Tue Aug 24 04:52:41 1999
From: pointal at lure.u-psud.fr (Laurent POINTAL)
Date: Tue, 24 Aug 1999 08:52:41 GMT
Subject: newbie: Why Py_Initialize( ) doesn't work in VC6.0
References: <7ptah7$hvl$1@newsource.ihug.co.nz>
Message-ID: <37c25cdf.93163762@news.u-psud.fr>

On Tue, 24 Aug 1999 17:24:18 +0800, "Bing Chen" <BC at prism.co.nz>
wrote:

>: Hi, all
>: I try to emed Python into my application, follow the manual setting the
>: project in VC6.0. In Object/library modules: python15.lib
>:                               in Additional include directories: d:\program
>: files\python\include
>:                              In Main program: #include "python.h"
>:                                                            Py_Initialize()
>: When I debug it, always show me "CXX0017:Error:symbol 'Py_Initialize' not
>: found.

And in "Link - Input - Additional library path:", do you set the
python libs path?

A+

Laurent.



From earlybird at mop.no  Sun Aug 29 19:25:01 1999
From: earlybird at mop.no (Alexander Staubo)
Date: Mon, 30 Aug 1999 01:25:01 +0200
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk>
Message-ID: <MPG.1233dfb6145766d998968c@news.mop.no>

In article <935279776snz at vision25.demon.co.uk>, 
philh at vision25.demon.co.uk says...
> 
> 
> It looks like version 0.0.1 of Parrot, my text-based GUI builder,
> will be ready within a day or so. It doesn't actually do anything
> useful yet; v0.0.1 is a technology demonstration version, which
> is intended to verify that the basic design is more-or-less
> sensible. Would anyone like to have a look at it, and possibly 
> suggest some improvements? If so, tell me, and I'll put it on
> my website.
> 
> Some information about Parrot follows:
[lots of stuff snipped]

I see no mention of two-way synchronization -- does this mean that once 
you have written your Parrot script and edited the generated source, 
your GUI is essentially locked?

The problem as I see it is that the descriptive language is not the 
primary tool: once you've generated your GUI source code (eg., in 
Python), you're separated from the original script. There's no way to 
re-generate the GUI framework and preserving your logic -- because there 
is no separation here between presentation and code. You assume they're 
be intermingled.

The issue is not just about good design, but also about practicalness: 
If you run Parrot, its backend will (I conjecture) simply overwrite the 
Python script, and your code will be lost. Unless each backend can 
correctly discern what code is generated, and what is added by the user, 
you'll have one hell of a time maintaining the source. Or perhaps you 
expect people to run the Parrot generator only once.

A more interesting approach would be to create a Python framework that 
loaded such Parrot scripts and instantiated necessary objects at 
runtime. (This could equally well be done through a multiple backend 
system, but once one has come this far, one might wonder whether a 
"backend" system would be relevant anyway.)

For a good model on presentation/code separation, examine the Delphi VCL 
framework. Your application is pure code; the rest is stored separately 
in plaintext asset files, in a proprietary syntax not unlike your Parrot 
language:

object MyForm: TMyForm
  Left = 191
  Top = 103
  Width = 333
  Height = 131
  Caption = 'Form1'
  Color = clBtnFace
  Font.Charset = DEFAULT_CHARSET
  Font.Color = clWindowText
  Font.Height = -11
  Font.Name = 'MS Sans Serif'
  Font.Style = []
  OldCreateOrder = False
  PixelsPerInch = 96
  TextHeight = 13
  object OkButton: TButton
    Left = 32
    Top = 48
    Width = 75
    Height = 23
    Caption = 'OK'
    TabOrder = 0
  end
  object MyEdit: TEdit
    Left = 32
    Top = 16
    Width = 209
    Height = 21
    TabOrder = 1
    Text = 'This is an edit field.'
  end
end

(It's worth pointing out that, although you can edit these scripts, as a 
rule you don't create these scripts yourself. They're generated by the 
visual GUI builder. Most people don't even see these scripts.)

The Delphi IDE itself has a neat two-way synchronization mechanism that 
allows you to edit the code and still remain 100% in sync with the GUI. 
For example, let's say you bind some code -- an "OnClick" event handler 
-- to a button. You can edit the button, move it around, or edit the 
event handler's code, move _that_ around -- but the link between the 
source code and the GUI description is never severed.

(The two-way engine actually uses its own parser. Because the language 
-- ObjectPascal -- is sufficiently strict, it can transparent re-link 
GUI-bound handlers by looking at their names and syntax. Contrast this 
with other popular GUI builders such as Visual C++ and, afaik, Visual 
Cafe, which use machine-readable comments to sync their source and GUI. 
You can't really blame these shortcomings Java or C++ -- Delphi comes in 
a nearly identical C++ version called C++Builder, which provides the 
exact same two-way feature.)

Please don't construe my comments as being anything but constructive 
criticism. I want great GUI tools for Python, too.

-- 
Alexander Staubo             http://www.mop.no/~alex/
"`Ford, you're turning into a penguin. Stop it.'"
--Douglas Adams, _The Hitchhiker's Guide to the Galaxy_



From aahz at netcom.com  Sun Aug  1 20:20:53 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 2 Aug 1999 00:20:53 GMT
Subject: Q: Why is __init__.py required
References: <7o2eeq$gs2@dfw-ixnews6.ix.netcom.com>
Message-ID: <7o2o95$eun@dfw-ixnews11.ix.netcom.com>

In article <7o2eeq$gs2 at dfw-ixnews6.ix.netcom.com>,
G. David Kuhlman <dkuhlman at netcom.com> wrote:
>
>I have looked and looked for this.  Whenever I put an empty __init__.py
>in a package, I ask myself why I have to do this.  I read Guido's
>article on packages and it told me that I have to put an __init__.py in
>a directory in order for the directory to be recognized as a package.
>But it doesn't say why.

Look at it this way: say you give Python the name of a directory.  How
does Python know that the directory really is a package, rather than
some random directory?  Like much else in Python (e.g. ":", "=="), this
straitjacket is designed to save you time by preventing stupid mistakes.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From kistler at fnmail.com  Mon Aug 23 05:13:37 1999
From: kistler at fnmail.com (Per Kistler)
Date: Mon, 23 Aug 1999 11:13:37 +0200
Subject: Runtime check of methods
Message-ID: <37C110C1.AA220DD0@fnmail.com>

Hi

Why does Python only runtime check, wether it knows
a method? For instance, if I forget to put the
"self." in front of a method in a class, than
python myclass.py still does not complain.

Thanks, Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From sp00fD at yahoo.com  Thu Aug 26 08:34:16 1999
From: sp00fD at yahoo.com (sp00fD)
Date: Thu, 26 Aug 1999 12:34:16 GMT
Subject: Documentation string
Message-ID: <7q3c88$k8m$1@nnrp1.deja.com>

Can someone tell me why the

def foo(self):
"""A method"""

won't work?

$ ./rcopy.py -p BACKUP ./foo /tmp
  File "./rcopy.py", line 18
    """A method"""
                 ^
SyntaxError: invalid syntax


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gtnorton at my-deja.com  Mon Aug 30 23:48:09 1999
From: gtnorton at my-deja.com (gtnorton at my-deja.com)
Date: Tue, 31 Aug 1999 03:48:09 GMT
Subject: Grids in Tkinter
References: <936050813snz@vision25.demon.co.uk>
Message-ID: <7qfj9m$rmo$1@nnrp1.deja.com>

In article <936050813snz at vision25.demon.co.uk>,
  philh at vision25.demon.co.uk wrote:
>
> Can anyone point me to anywhere that documents how Grids work in
> Tkinter? A short example program would be nice.
>
> --
> Phil Hunt....philh at vision25.demon.co.uk
>
>
I'm sure Fredrik Lundh's Introduction to Tkinter has what your looking
for:             www.pythonware.com/library.htm

                        A Great Read!
                                     Good Luck,
                                              gtnorton


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From puetsch at gmx.de  Tue Aug  3 13:21:47 1999
From: puetsch at gmx.de (Felix Puetsch)
Date: 3 Aug 1999 17:21:47 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk>
Message-ID: <7o78fb$blb$2@chief.death.de>

I.Clarke at strs.co.uk (Ian Clarke) wrote in
<37A6A2DB.9B11FB01 at strs.co.uk>: 

>It is to create a Python web server which automatically parses
>embedded Python in HTML (there is a Perl extension that does this
>too).  So, for example, the following HTML code:
>
><HTML>
><HEAD>
><TITLE>
>{
>  n = 10
>  print n*2
>}
></TITLE>
></HEAD>
><BODY>
>{
>  print n*3
>}
></BODY>
></HTML>
>

How would you make a web page display this message, e.g.? I think you 
should at least define environments for Python code. ('<!--' etc.) Do 
you know PHP3?

-- 
Felix P?tsch



From meowing at banet.net  Mon Aug 23 13:34:02 1999
From: meowing at banet.net (meow)
Date: 23 Aug 1999 13:34:02 -0400
Subject: Runtime check of methods
References: <37C110C1.AA220DD0@fnmail.com>
Message-ID: <87u2pqo0z9.fsf@slip-32-100-244-100.ma.us.ibm.net>

Per Kistler <kistler at fnmail.com> wrote:

> Why does Python only runtime check, wether it knows a method? For
> instance, if I forget to put the "self." in front of a method in a
> class, than python myclass.py still does not complain.

It's not necessarily an error.  Remember that this language treats
everything as a big pile of function pointers and dispatch tables, and
you can mess around with them at run time.  Try this:

=-=-=-= goofymod.py =-=-=-=
class Goofyclass:

   def doit(self):
       print unknownsub()    # This doesn't look at all healthy.
#eof

=-=-=-= main.py =-=-=-=
import goofymod

def useful():
    return "meow"

foo = goofymod.Goofyclass()  # We still have no unknownsub,
goofymod.unknownsub = useful # but fix that here,
foo.doit()                   # and it even works.
#eof



From I.Clarke at strs.co.uk  Mon Aug 23 08:26:33 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Mon, 23 Aug 1999 13:26:33 +0100
Subject: Python / Cool looking applications?
References: <m3ogfz9mx6.fsf@intergate.bc.ca>
Message-ID: <37C13DF9.5400480@strs.co.uk>

Depends on the platform I guess.  Python can do [pretty much] anything
that TK can do on either Windows or Unix, so just look for cool-looking
TK applications to get an idea.

If you are using Linux, you can use pyGTK - GTK is, I think without a
doubt, the coolest looking Toolkit around (just checkout
http://gtk.themes.org/).

Ian.

Gerald Gutierrez wrote:
> 
> Hi all.
> 
> I've been looking into using Python as UI glue and at Python's imaging
> extensions. Are there any examples of very cool looking interfaces or
> otherwise well-designed interfaces that are built in Python?
> 
> Also, does anyone know of any applications that use Python's imaging
> extensions?
> 
> Thanks very much.
> 
> Please forward replies to gutier at intergate.bc.ca.



From Sven.Drescher at dlr.de  Tue Aug 31 10:25:58 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Tue, 31 Aug 1999 16:25:58 +0200
Subject: C++ and Python
Message-ID: <7qgnr5$ls8@news.go.dlr.de>

Hi!

I have a little problem. It's really urgent. I wrote a program with threads.
One thread is the Python/Tk loop with my graphical user interface and the
other is an work thread.

Now my question:
How can I get the entrie-variables of python in my work thread.

I know such functions like PyArg_ParseTuple and so on to convert the values,
but is there a possibility to have an C-reference variable to a python
variable or so?

Has anyone done this before?

Thanks for hints and fact!

Sven
______________________________________
German Aerospace Research Establishment






From amk at mira.erols.com  Sun Aug 15 21:51:54 1999
From: amk at mira.erols.com (A.M. Kuchling)
Date: 16 Aug 1999 01:51:54 GMT
Subject: Question: copy/deepcopy with the curses module?
References: <0eCt3.1775$kW2.100172@typhoon2.gnilink.net>
Message-ID: <slrn7res4l.3ma.amk@207-172-146-60.s60.tnt3.ann.va.dialup.rcn.com>

On Sun, 15 Aug 1999 16:51:08 GMT, Damond Walker <dwalker at iximd.com> wrote:
>    What I want to do is get a snapshot of the current SIO instance.  This
>would allow me to quickly save the current SIO instance in a temp. variable,
>do some screen updates, and then quickly repaint the screen by restoring the
>SIO instance to it's older value.

Here's a suggestion (that I haven't tried): create a new window the
size of the entire screen and render the SIO instance into it; this
should overwrite all the other information on the screen.  Then
restore the old display by deleting the new window and refreshing the
screen.  Something like 'win = stdscr.newwin(0,0) ; ... do stuff with
win ; del win ; stdscr.refresh()'...

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
The garden of Destiny. Look behind you: shadow-plays of memory are forever
being enacted, on paths you walked too long ago.
    -- From SANDMAN #47: "Brief Lives:7"



From chapman at bioreason.com  Fri Aug  6 10:40:21 1999
From: chapman at bioreason.com (Mitch Chapman)
Date: Fri, 06 Aug 1999 08:40:21 -0600
Subject: Linux why not fully dynamic?
References: <1278177653-25589496@hypernet.com>
Message-ID: <37AAF3D5.141B55EB@bioreason.com>

Gordon McMillan wrote:
> 
> Robin Becker asks:
> 
> > Why doesn't ./configure set up for a libpython.so.1.5.2 or whatever?
> 
> Don't know. Maybe because this kind of scheme doesn't work on
> some other Unixen?

This question (or something similar to it) is answered in the Python
FAQ:
	http://www.python.org/doc/FAQ.html#3.30

-----------------------------------------------------------------------
3.30. Why is the Python interpreter not built as a shared
library?

(This is a Unix question; on Mac and Windows, it is a shared library.) 

It's just a nightmare to get this to work on all different
platforms. Shared library portability is a pain. And yes, I know about
GNU libtool -- but it requires me to use its conventions for filenames
etc, and it would require a complete and utter rewrite of all the
makefile and config tools I'm currently using.

In practice, few applications embed Python -- it's much more common to
have Python extensions, which already are shared libraries. Also,
serious embedders often want total control over which Python version
and configuration they use so they wouldn't want to use a standard
shared library anyway. So while the motivation of saving space when
lots of apps embed Python is nice in theory, I doubt that it will save
much in practice.  (Hence the low priority I give to making a shared
library.)
-----------------------------------------------------------------------

-- 
Mitch Chapman
chapman at bioreason.com



From mcfletch at vrtelecom.com  Wed Aug 11 14:03:53 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Wed, 11 Aug 1999 18:03:53 GMT
Subject: Why is CGI module so complicated?
In-Reply-To: <cWYr3.1329$EG4.226668@news.shore.net>
References: <cWYr3.1329$EG4.226668@news.shore.net>
Message-ID: <NDBBIAPHEKGOBIHCBLIIMEDNCAAA.mcfletch@vrtelecom.com>

What about simply adding a __str__ method to these objects?

That way for printing and '%s'%(object) substitution you get the "simple"
response of returning the (likely de-escaped) string value stored in the
field. Similarly, add a "list" method to get the values as a list (possibly
of length one) (is there a special method name that the built-in list
function calls? I could not find it.)

Disclaimer:
	I do not use CGI at all anymore, no idea if this even make sense for
day-to-day operations.

Enjoy yourselves,
Mike


-----Original Message-----
From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
Behalf Of Michael P. Reilly
Sent: August 10, 1999 1:02 PM
To: python-list at cwi.nl
Subject: Re: Why is CGI module so complicated?


Ian Clarke <I.Clarke at NOSPAM.strs.co.uk> wrote:
: I have been writing an application using the cgi module, and am somewhat
: confused as to why data is returned in FieldStorage or MiniField storage
: objects, why not just return it in a plain dictionary?

: At the moment, lets say you want to access a CGI variable called "pie"
: you need to use:

: f = cgi.FieldStorage()
: print f['pie'].value

: But why not just:

: f = cgi.FieldStorage()
: print f['pie']

: That way the FieldStorage object would be a true dictionary, and could
: be used as, for example, a TemplateDocument's substitutions dictionary.

: I would write a wrapper to get rid of all of the FieldStorage objects,
: but I have almost finished my code :-(

: Can anyone explain what the point of the FieldStorage and
: MiniFieldStorage objects are?

: Ian

Ian, please read my reply in the previous thread:
  http://www.dejanews.com/getdoc.xp/AN=502365247

  -Arcege





From ramune at bigfoot.com  Tue Aug 24 16:04:42 1999
From: ramune at bigfoot.com (Graffiti)
Date: 24 Aug 1999 13:04:42 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net> <slrn7s2lb3.k6b.abigail@alexandra.delanet.com>
Message-ID: <7putsq$3ql$1@lycaeus.calstatela.edu>

In article <slrn7s2lb3.k6b.abigail at alexandra.delanet.com>,
Abigail <abigail at delanet.com> wrote:
[snip amusing rant]
>Yeah, those math texts from the 16th and 17th century, they are sooooo easy
>to read, because they lack all the symbols. Not to mention the original
>Euclid and those other Greeks, who had no symbols at all!

<GRIN>
Oh, but they're written with only symbols!  All those confusing greek symbols
that mathematicians use are the basic building blocks for the language!  Why
didn't they just use English?
</GRIN>

>"" On a personal note...  Do you know what a computer science degree is these
>"" days? They require you understand shits like lambda calculus, logic, graph
>"" theory, combinatorics, or even algorithms?? All for what? All the Perl

Sure, why not?

>"" programers I know doesn't understand any of these, and they all are making
>"" big bucks in fortune 500. I'm not ashamed that I dropped out of school. I
>"" hope that Larry and Tom's teachings will eventually make such subjects
>"" disappear for good.

You don't *have* to know them.  Just in the same way you don't *have* to
Perl to get a job administering web servers.  They just help.  (Of course,
while the abstract material in CS can come in handy, if they're taught
wrong, what's the point?)

>Hahahahahahahahahahhahahahahaahahahahahaha.
>
>
>If you think there's much relationship between programming and Computer
>Science, you're utterly mistaken. It's the same relationship between
>driving your Morris Minor and designing and producing a new car.

There's quite a bit of relation.  You don't need to understand how relational
databases work and the theory behind them.  But I wouldn't trust my
business or life to a database server written by someone who didn't have a
firm understanding of database theory.

Same way, I wouldn't want someone who never studied lambda calculus and logic
to write an expert system that would be used for emergency room diagnostics
where the wrong decision can KILL.  (I've heard of Prolog used to write an
expert system for just this purpose.  Dunno about perl.)

Just think of them as a way to increase your understanding.  (Of course,
whether the instructors at most universities are competent and know what
they're teaching is a different story altogether. :-)

>Larry and Tom can teach all they want; their teachings will have no impact
>on Computer Science what so ever. Just like their teaching will have no
>impact on the grow of potatoes in Idaho.

Sure it will.  Much of what they teach will expose Perl programmers to new
ways of thinking and solving problems.  By helping people think out of the
electrified-ivory-tower-deathmatch-cage that some schools tend to put up
around the students' minds, they'll help people find new and innovative
ways to solve problems.

And it's not so far-fetched to think that agricultural supply vendors would
be using perl to help manage operations for their clients, which may include
potato farmers, no? ;-)

-- DN



From zontar4 at yahoo.com  Fri Aug 27 04:28:12 1999
From: zontar4 at yahoo.com (G. Lewis)
Date: Fri, 27 Aug 1999 08:28:12 GMT
Subject: idling with tkinter
References: <7q2u6e$aso$1@nnrp1.deja.com> <7q357r$c50c@eGroups.com>
Message-ID: <7q5i6j$5qu$1@nnrp1.deja.com>

In article <7q357r$c50c at eGroups.com>,
  john.grayson at gsc.gte.com wrote:
> no pending events in the event queue; you must setup to be called
> *again* before you return from the callback (this is true for after,

Thanks john and fredrick for the tips... very helpful.

--
...


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From skip at mojam.com  Thu Aug 26 11:47:52 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 10:47:52 -0500 (CDT)
Subject: Newbie: About PyRun_File()
In-Reply-To: <7q2mn1$sik$1@newsource.ihug.co.nz>
References: <7q2mn1$sik$1@newsource.ihug.co.nz>
Message-ID: <14277.24954.983888.279871@dolphin.mojam.com>

    Bing> How can I use PyRun_File() or PyRun_AnyFile() in my application to run a
    Bing> Python file? or where can I got an example?

Try the extending and embedding manual:

    http://www.python.org/doc/current/ext/ext.html

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...




From hyoon at bigfoot.com  Mon Aug 16 16:33:32 1999
From: hyoon at bigfoot.com (Hoon Yoon)
Date: Mon, 16 Aug 1999 16:33:32 -0400
Subject: Bug in struct module?
Message-ID: <37B8759C.D0DF1B55@bigfoot.com>

Hi,


>>> calcsize('hd')
16
>>> calcsize('dh')
10
That seems like a bug to me. BTW: it won't convert 
>>> unpack('dh', s[-10:])
(2.47032822921e-323, 16424)
>>> unpack('hd', s[-10:])
Traceback (innermost last):
  File "<stdin>", line 1, in ?
struct.error: unpack str size does not match format

Does anyone know how to unpack a bin file written in VB:
Fno = FreeFile
Open CDir & Fn & "." & cdte For Binary As #Fno
Put #Fno, , xMat

where xMat is same as x created by?

Dim x(4, 3) As Double
'This code example opens the file in Binary mode for reading; other
processes can't read file.
For r = 0 To UBound(x, 1)
    For c = 0 To UBound(x, 2)
        x(r, c) = r * c
    Next c
Next r

-- 
*****************************************************************************
S. Hoon Yoon (Quant) Merrill Lynch Equity Trading, 
yelled at yahoo.com hoon at bigfoot.com(w)
"Miracle is always only few standard deviations away, but so is
catastrophe."
* Expressed opinions are often my own, but NOT my employer's.
"I feel like a fugitive from the law of averages."    Mauldin
*****************************************************************************
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vcard.vcf
Type: text/x-vcard
Size: 202 bytes
Desc: Card for Hoon Yoon
URL: <http://mail.python.org/pipermail/python-list/attachments/19990816/82b8226a/attachment.vcf>

From aahz at netcom.com  Mon Aug 30 14:11:14 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 30 Aug 1999 18:11:14 GMT
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net>
Message-ID: <7qehg2$bkc@dfw-ixnews4.ix.netcom.com>

In article <Pine.LNX.4.10.9908281230060.9521-100000 at starship.python.net>,
Moshe Zadka  <moshez at server.python.net> wrote:
>On Sat, 28 Aug 1999, Tim Peters wrote:
>>
>> python-folk-are-the-best-buddies-psu-dues-can-buy-ly y'rs  - tim
>                                   ^^^
>There are a few options:
>
>1. The mailman at Python org had a glitch
>2. The timbot made a type
>3. There's this underground association, Python Software Union, which goes
>around breaking thumbs of people who use Python without paying the PSA.

Sheesh, does nobody here remember psuvm?
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From bruce_dodson at bigfoot.com  Fri Aug  6 13:25:56 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Fri, 6 Aug 1999 14:25:56 -0300
Subject: Any clues to source of this delay?
References: <4D0A23B3F74DD111ACCD00805F31D8100DB91039@RED-MSG-50>
Message-ID: <KYEq3.51226$jl.34300060@newscontent-01.sprint.ca>

>Windows does indeed support this socket option.


I didn't say that it did not.  Take a closer look at my example code:

>>   if (os.name == 'nt') ...





From skaller at maxtal.com.au  Sun Aug 15 13:31:36 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sun, 15 Aug 1999 17:31:36 GMT
Subject: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <q43s3.58867$jl.37270547@newscontent-01.sprint.ca> <934412029.374060@news.adtranzsig.de>
Message-ID: <37b6f6a4.349078969@news.triode.net.au>

On Thu, 12 Aug 1999 00:15:12 +0200, "Dirk-Ulrich Heise" <hei at adtranzsig.de> wrote:

>Bruce Dodson schrieb in Nachricht ...
>>Ni!  I don't see the point of creating a variant of python that aims to be
>>more efficient, but does so at the cost of taking away Python's most
>>powerful optimization mechanism: external C modules.

	The answer is simple: external C modules
may be the most powerful optimisation mechanism _at present_,
but with a compiler, that will no longer be true.

	'foreign' modules will still be needed to add functionality
that cannot be implemented in Python, such as access to OS services.
However, if pure python can be compiled to efficient code,
there is no need to hassle with C modules just to get better performance.

>Here, i agree with Bruce.
>At the moment, it's common practice to accelerate a Python
>program that's too slow by moving parts of it into homebrew
>extension modules. Taking away this option would maybe
>invalidate Viper as a means of optimization. 

	Why? If you _kept_ the original python
source from which the C code is derived, then you
could just throw out the C code and use the Python code
again. If you didn't, you are in for trouble when the Python
C API is upgraded or changed, or you decide to mode
to JPython.

>It would be okay
>with me, though, if you allow a different kind of extension
>modules (a different interface to the extension module)-
>if ViperC creates C code, i don't see a big problem with
>that.

	There will be at least one, or perhaps more,
ways of adding extensions, but the primary purpose of this
will be

	a) for compatibility (not efficiency)
	b) for adding new functionality that pure python cannot support

There is a secondary purpose in allowing foreign language extensions
for efficiency, when Python does not have the constructions to
allow the compiler to optimise the code (or it is simply too hard to
do so).

	In these cases, modifying the compiler, or adding
extensions to the language, may prove a better option,
since it retains the 'pure language' approach. In these
cases backwards compatability is lost, but it is also
true whenever you write a C module, you are bound to
a particular C API and lose compatibility with JPython
and probably CPython 1.6 and 2.0 and 1.4, etc.

>In this case, i disagree with Bruce. Implementing a subset
>of Pythons semantic possibilities and keeping the same syntax makes
>sense in real life. Just try to write your production code in a less
>wild fashion than usually, not changing bindings all over, and you'd have
>a comfortable way of getting at more speed without rewriting as
>much stuff as if you had to port it to C.

	Yes, that is the basic premise: you can use Python's
nice syntax, and get _efficient_ code, and if you are careful,
it will run on CPython, JPython, and Viper. 

	This is the kind of code that one should aim for
most of the time anyhow: if you really do use nasty tricks,
at least the use should be isolated, and the dependence
reasonably transparent.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From tim_one at email.msn.com  Tue Aug 31 01:44:20 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 31 Aug 1999 01:44:20 -0400
Subject: Porting Python
In-Reply-To: <37ca6986@nntp.server.uni-frankfurt.de>
Message-ID: <000201bef373$df6ffe80$932d153f@tim>

[mlauer at trollinger-fe.rz.uni-frankfurt.de]
> While we're at the subject of porting, does anyone
> know of a port to the Psion Series 5 ?

    http://dales.rmplc.co.uk:9080/Duncan/PyPsion.htm





From dl231 at eng.cam.ac.uk  Thu Aug  5 12:34:52 1999
From: dl231 at eng.cam.ac.uk (D. Lopez-De-Ipina)
Date: Thu, 05 Aug 1999 17:34:52 +0100
Subject: CORBA IDL type objects serialization problems
Message-ID: <37A9BD2C.DA6C5580@eng.cam.ac.uk>

Hi,

I am trying to store CORBA IDL type objects in a Python shelve. I am
using Fnorb ORB. I managed to serialize CORBA.string type values but
when I try with other simple built-in types such as short, the program
crashes. I have included the code that doesn't work and the resulting
stack trace. I wonder if anybody could please explain why CORBA.string
IDL type can be serialized and not the others? If it is possible and
I have just done something wrong, would if be possible to serialize user
defined IDL types on a server without this one having knowledge of the
IDL type trasferred within an Any?

Very grateful,

Diego

#!/usr/bin/env python

import shelve

# Fnorb modules.
from Fnorb.orb import CORBA

if __name__ == '__main__':
  dictionary = shelve.open('dummy')
  dictionary['str'] = CORBA.Any(CORBA.TC_string, 'hello')
  print "dictionary['str']:", dictionary['str'].value()

  dictionary['short'] = CORBA.Any(CORBA.TC_short, 5)
  print "dictionary['short']:", dictionary['short'].value()

[dl231 at worthingtons DynAny]$ python prueba.py
dictionary['str']: hello
dictionary['short']:
Traceback (innermost last):
  File "prueba.py", line 14, in ?
    print "dictionary['short']:", dictionary['short'].value()
  File "/lce/lib/python1.5/shelve.py", line 65, in __getitem__
    return Unpickler(f).load()
TypeError: ('too many arguments; expected 1, got 2', <class
Fnorb.orb.TypeCode.ShortTypeCode at 8170908>, (tk_short,))





From jdnier at execpc.com  Mon Aug  9 18:10:46 1999
From: jdnier at execpc.com (David Niergarth)
Date: Mon, 9 Aug 1999 17:10:46 -0500
Subject: win32api -- long vs short filenames
References: <7oa0n0$72n@newsops.execpc.com>
Message-ID: <7onjk9$qr2@newsops.execpc.com>

David Niergarth <jdnier at execpc.com> wrote in message
news:7oa0n0$72n at newsops.execpc.com...
> I'm trying to expand a mangled DOS 8+3 name (from WinNT4.0) into the long
> version of the name. There's a win32api function for doing the opposite,
for
> example,
>
> >>> win32api.GetShortPathName(r'c:\long_path_name\long_file_name.txt')
> 'c:\\LONG_P~1\\LONG_F~1.TXT'
>
> But I can't figure out how to expand a mangled name into the original long
> name. Any ideas?
>
> --David Niergarth

David Asher (thanks, by the way) pointed me to win32api.FindFiles(), as in

>>> win32api.FindFiles(r'c:\PROGRA~1')[0][-2]
'Program Files'

This returns the long version of the last component of the path that you
feed it but not the long version of the entire path, which is what I'm
after. In case anybody is interested, here's a function that will return the
long version of the entire path from either a mangled absolute path or a
mangled simple file name in Python's current directory.

def win32_GetLongPathName(pathname):
    """Transform a mangled 8+3 pathname to original long name.

    pathname -- Can be an absolute path or a simple file or direcotry name
                if it exists in Python's notion of the current directory.

    If a simple file or directory name cannot be resolved or an absolute
    path cannot be found, the original pathname will be silently returned.

    """
    import os, win32api
    if os.path.isabs(pathname):
        abspath = pathname
    else:
        cur_dir = os.getcwd()
        newpath = os.path.join(cur_dir, pathname)
        if os.path.isfile(newpath):
            abspath = newpath
        else:
            return pathname  # could not resolve relative path
    pieces = []
    head, tail = os.path.split(abspath)
    while tail:
        try:
            longtail = win32api.FindFiles(os.path.join(head, tail))[0][-2]
        except IndexError:
            return pathname  # could not find path
        pieces.insert(0, longtail)
        head, tail = os.path.split(head)
    pieces.insert(0, head)
    return apply(os.path.join, pieces)

Seems like there should be an easier way, but this works for me.

--David Niergarth






From claird at starbase.neosoft.com  Sun Aug 22 08:47:27 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Aug 1999 12:47:27 GMT
Subject: Setting environment variables.
References: <37BD3429.7364B4F7@iinet.net.au> <37BDB88C.212EC063@starmedia.net>
Message-ID: <9365FFF976D270AA.74059ADD92F004AE.52007F4DA274524D@lp.airnews.net>

In article <37BDB88C.212EC063 at starmedia.net>,
Andrew Csillag  <andrew at starmedia.net> wrote:
>Oliver White wrote:
			.
			.
			.
>> I'm running GNU/Linux, is there an issue with processes... eg, after
>> running my python program as a child process, will the changes to the
>> environment variable affect it's parent (the *term)? (suspecting no) Is
>> there annother way to set the parent's?
>
>You would be correct, changing environment variables in a child process
>won't affect the parents environment variables.  A way to do it is have
>the child process print the new value of the environment variable to
>stdout and have the parent read this value (probably using os.popen or
>similar) and set its environment accordingly.
			.
			.
			.
What's traditional for Unix programmers working within
{sh,bash,ksh,...} is to
  eval `my_env_setting_app`
at the command prompt.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From aahz at netcom.com  Tue Aug 24 13:34:32 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 24 Aug 1999 17:34:32 GMT
Subject: Converting from perl
References: <7pujub$75v$1@nnrp1.deja.com>
Message-ID: <7pul38$5an@dfw-ixnews17.ix.netcom.com>

In article <7pujub$75v$1 at nnrp1.deja.com>, sp00fD  <sp00fD at yahoo.com> wrote:
>
>I'm converting to python from perl and just created my first python
>program, it's actually a rewrite of a perl program that I had.  

Welcome to comp.lang.py.

It would help if you explained what your script is *intended* to do.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From philh at vision25.demon.co.uk  Sun Aug  8 07:26:09 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 08 Aug 99 11:26:09 GMT
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk> <slrn7qq4as.cc2.kas@titan.magnetic-ink.dk>
Message-ID: <934111569snz@vision25.demon.co.uk>

In article <slrn7qq4as.cc2.kas at titan.magnetic-ink.dk>
           kas at maps.magnetic-ink.dk "Klaus Alexander Seistrup" writes:
> Phil Hunt wrote:
> >    #bak.py
> >    from hhh import *
> >
> > [...]
> > 
> >    #hhh.py
> >    from bak import *
> 
> Do we have a circular reference here?

Yes. Does Python not allow this? What's the workaround? (Originally 
my two modules were one, with hhh containing subclasses of bak's
classes, and bak containing code to dynamically create instances
of hhh classes at run time, using:

   newInstance = eval(nameOfClass + '()')

but I wanted the code split up into two modules as it is logically 
separate.)

-- 
Phil Hunt....philh at vision25.demon.co.uk




From JamesL at Lugoj.Com  Sun Aug 15 17:06:04 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Sun, 15 Aug 1999 14:06:04 -0700
Subject: Peace in our time? Alas no. Was: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com>
Message-ID: <37B72BBC.2172F66C@Lugoj.Com>

(The problem with these language wars is that there is no way for a
self-respecting arms merchant to sell to both sides; there just isn't
anything to sell.)

I suspect that computer language wars arise for some of the same reasons
some real wars arise: a battle over finite resources. In this case, the
finite resource is a programmer's time. I know I don't have time to find the
"Holy Grail" (ahem, sorry!) of languages by learning every language ever
invented. Once I've somehow settled on a language and spent an irretrievable
fraction of my life learning it, I get irritable when someone tells me I
have to do that process over again. For me, the interest in computer
languages per se passed a decade and a half ago. I've learned and glanced at
(and forgotten) far too many languages and notations (at least 22, probably
more). It is a tiny subset of all the languages that have been commercially
released, but to my mind is too bloody much. Bottom line for me (and I
suspect for others) is: Please don't ask me to change from X to Y unless you
have a damn good reason; my time is precious.

But sometimes people have to discard something that works perfectly fine and
learn another language, simply because they move into a group that mandates
the use of something else. Then the irritation and fight over languages
happens. There is no avoiding it I think, so these languages wars will go on
for as long as there are multiple languages (and anyone who thinks that will
change is dreaming).



From news at dorb.com  Sun Aug  8 12:21:49 1999
From: news at dorb.com (Darrell)
Date: Sun, 8 Aug 1999 12:21:49 -0400
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk>
Message-ID: <N8ir3.937$zN2.21156@typhoon.nyroc.rr.com>

I think this is one of those things I had to forget about from
the world of 'C'. In Python you don't need to have an object
defined by some .H file to use it. Besides circular reverences
importing this way isn't much different than having one large
module.

As you know one large module isn't a nice thing.

Try something like passing a parm that allows access to
an object in another module. The parameter approach also
allows you to change the relationships of these modules
at runtime.

Decoupling from other modules is a good thing.

--
--Darrell

Phil Hunt <philh at vision25.demon.co.uk> wrote in message
news:934075710snz at vision25.demon.co.uk...
> I have two python modules, both of which need to refer to classes
> defined in the other one. How do I do this?
>
> I am currently putting at the start of each module a line of the form:
>
>    from theOtherModule import *
>
> However, this is not working. My modules -- greatly simplified -- look
> like this:
>
>    #bak.py
>    from hhh import *
>    ti = AnHhhClass()
>    print ti
>
> and:
>
>    #hhh.py
>    from bak import *
>    class AnHhhClass: pass
>
> When I run ``python bak.py'', I get the error message
> ``NameError: AnHhhClass''.
>
> --
> Phil Hunt....philh at vision25.demon.co.uk
>





From glen at ranma.interq.or.jp  Thu Aug 19 21:57:30 1999
From: glen at ranma.interq.or.jp (Glen Malley)
Date: 20 Aug 1999 10:57:30 +0900
Subject: Class Description
References: <37ba0d28@news.interq.or.jp> <37BAA047.275A1536@strs.co.uk>
Message-ID: <37bcb60a@news.interq.or.jp>

On Wed, 18 Aug 1999 13:00:07 +0100, Ian Clarke <I.Clarke at strs.co.uk> wrote:
>
>> ...could a class re-write itself, if properly messed with?
>
>Do you mean like adding and removing methods and fields?

I mean like re-writing itself.
Adding and removing methods and fields is just one part
of it. I'd like to make something which can modify its
own source based on the additions made to it.
Shouldn't be too hard so long as classes can describe themselves.




From che at debian.org  Tue Aug 10 15:12:43 1999
From: che at debian.org (Ben Gertzfield)
Date: 10 Aug 1999 12:12:43 -0700
Subject: posix.error is a tuple, not an object or a string as docs say
References: <yttzozzbm11.fsf@gilgamesh.cse.ucsc.edu> <KM_r3.1361$EG4.234560@news.shore.net>
Message-ID: <yttu2q7bgb8.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Michael" == Michael P Reilly <arcege at shore.net> writes:

    Michael> The long and short of this is that you can use it like a
    Michael> tuple if more than one argument was given, and as the
    Michael> object otherwise.

    Michael> For some of your own reassurance, print the type of
    Michael> "error" in the except clause.

Right, but my problem is that I can't use it as an object as the
documentation says..

Python 1.5.1 (#1, Dec 17 1998, 20:58:15)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> try:
...     input = os.fdopen(42)
... except os.error, error:
...     print error.errno
...     print error.strerror
... 
Traceback (innermost last):
  File "<stdin>", line 4, in ?
AttributeError: errno

Yet the documentation says:

"When exceptions are classes, this exception carries two attributes,
errno and strerror. The first holds the value of the C errno variable,
and the latter holds the corresponding error message from strerror()."

So something else is a bit wacky, no?

-- 
Brought to you by the letters C and H and the number 11.
"Ha ha! I have evaded you with the aid of these pasty white mints!"
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From charles.choi at computer.org  Thu Aug 12 05:35:31 1999
From: charles.choi at computer.org (Charles Y. Choi)
Date: Thu, 12 Aug 1999 02:35:31 -0700
Subject: converting Netscrape mailbox to Unix mbox format
References: <14256.35316.328575.118034@lrz.de>
			<m3wvv24wcg.fsf@atrus.jesus.cam.ac.uk> <14257.45321.128760.603105@lrz.de> <37B1BD55.FA34A1CC@callware.com>
Message-ID: <37B29563.BB8C978A@computer.org>


Ivan Van Laningham wrote:

> Hi All--
>
> Eugene Leitl wrote:
> >
> > Michael Hudson writes:
> >
> >  > Last I looked Netscape mailbox format *was* unix mbox format (on unix
> >  > anyway; can't speak for Netscape on windows) - have you tried pointing
> >  > pine at them?
> >
> > Yes. Both pine and VM. Both see a giant 5 MByte message.

Try using dos2unix (if you have it) on your Netscape mailbox.  That does the
trick for me.


-Charles






From ivanlan at callware.com  Thu Aug 26 12:04:44 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Thu, 26 Aug 1999 16:04:44 GMT
Subject: A Plea For A Return To Sanity!!!
References: <37bc1339@cs.colorado.edu>
		<37bc313b@cs.colorado.edu>
		<14268.17944.323020.242210@bitdiddle.cnri.reston.va.us>
		<019301beeaea$004c4d80$f29b12c2@secret.pythonware.com>
		<37bd4876@cs.colorado.edu>
		<MPG.1229cd5eed750a81989687@news.mop.no> <37C55649.8320449D@megsinet.net>
Message-ID: <37C5659C.8892E52@callware.com>

Hi All--

David Oppenheimer wrote:
> 
> Dear Group,
> 
> How we have let the ravings of a few PERL programmers postings shift the
> entire tide of this newsgroup to comparing Python & Perl and throwing various
> insults I do not know....Can we stop this now and get back to discussing how
> to better program in Python or am I misinterpreting the real reason for the
> existence of this group?!
> 

1)  If the Perl yahoos would quit cross-posting flaming bags of shit,
it'd be over.
2)  Now that Tom C has left for more fertile pastures, we have been
visited with Mr. X.  I wonder if we have 5 more visitations to go?
3)  The real reason for this newsgroup/mailing list is, ipso facto, the
discussion of Python.  I don't care for having to put on my asbestos
bunny suit just to read my mail.
4)  I use Netscape mail.  Can someone out there tell me how to killfile
Tom Christianson and Xah?  Thank you, and I apologize for the off-topic
nature of this request.  Or is it?


> Alexander Staubo wrote:
> 
> > "QED?" said Russell.
> > "It's Latin," said Morgan. "It means, 'So there you bastard'."
> > --Robert Rankin, _Nostramadus Ate My Hamster_

<i-wonder-what-he-shat>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From khalsa at acay.com.au  Wed Aug 25 06:46:19 1999
From: khalsa at acay.com.au (Gurvijay Singh Bhatti)
Date: Wed, 25 Aug 1999 20:46:19 +1000
Subject: equivalent of getch
Message-ID: <37C3C97A.59BF653E@acay.com.au>

hello everyone,

could someone tell me what is equivalent of getch in jpython? I want the
program to wait for the user to hit enter before going forward.

regards

GSB




From hnowak at cuci.nl  Thu Aug 26 13:44:56 1999
From: hnowak at cuci.nl (Hans Nowak)
Date: Thu, 26 Aug 1999 17:44:56 GMT
Subject: Fuzzy string matching?
Message-ID: <199908261746.TAA25440@dionysus.fw.cuci.nl>

...Anyone know of such a beast in Python?

The background is this... I want to write a deck managing program for 
Magic the Gathering (or other CCGs or similar games). The program has 
a collection of cards it knows (ideally, all kinds of cards in 
existance). Now I want people to enter decks, by entering names of 
cards. However, since some card names are quite difficult or have 
funny characters in them, I expect that some of their input will not 
be recognized in the card database. In such cases, I would like to 
present the user with a list of card names which match the input 
closely, but not entirely... strings that 'almost match'. OK, so it's 
just a fuzzy string match. Is such a thing done in Python already? If 
not, would it be difficult to do? (I obviously don't how to do it 
myself otherwise I would have done so already... :-/  )

TIA,

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/



From phd at emerald.netskate.ru  Sun Aug 29 05:56:18 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Sun, 29 Aug 1999 13:56:18 +0400 (MSD)
Subject: Loading shared modules from an imbedded interpreter
In-Reply-To: <37C84727.4E1EB9AF@delanet.com>
Message-ID: <Pine.LNX.4.04.9908291351480.10225-100000@emerald.netskate.ru>

On Sat, 28 Aug 1999, Dave Wallace wrote:
>     I am attempting to embed Python in a c executable, unfortunately
> when I attempt to import a shared-c module or a Python module that
> imports a shared-c module, I get errors like:
> 
>   Traceback (innermost last):
>   File "<string>", line 1, in ?
> ImportError: /usr/local/lib/python1.5/lib-dynload/selectmodule.so:
> undefined symbol: _Py_NoneStruct
> 
> where the undefined symbol varies from file to file. Apparently, when
> compiling parts of libpython are left out since they are not referenced
> by my code. Is there any way to force them to be included?
> 
> Thanks,
> Dave
> dwallace at netgsi.com

   You need to export these symbols from main program back to extensions.
The actual linker option is compiler/linker dependent. If you use gcc give
it the -Wl,-E option. With this gcc pass -E option down to linker.
   "man ld" for detailed technical description on -E.

   I stuck the problem when compiled PyApache - I can't load an extension.
PyApache mailing list gave me the advice on using -E - and now all go well!

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From jblaine at shell2.shore.net  Thu Aug 19 11:47:08 1999
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Thu, 19 Aug 1999 15:47:08 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
Message-ID: <slrn7ro9ns.h6v.jblaine@shell2.shore.net>

>Some have mentioned that I've left things out that are in
>more recent releases.  I'm sorry, but I only have Mark Lutz's
>book to go on.  There are no manpages for Python.   Like most
>people, I'll wait until the next edition of the book is release
>rather than digging around elsewhere if the manpage are missing.

Coming from a strong UNIX and Perl background (probably not as strong
as you...), I *never* once used a Perl man page throughout
writing tens of thousands of lines of code.  For that matter, I
never used the Tcl/Tk ones either.  I've never considered man pages
to be the Single Right Way First Line of Documentation Offering When
Documenting a Language.

So, maybe you use man pages all the time with Perl.  It is not
at all safe to assume that it's the primary means through which
people using Perl seek reference though (unless you've done some
sort of study I'm unaware of).

If you put me on any UNIX machine without man pages containing
descriptions and command-line usage of the tools there, I would
be royally pissed.

I don't know, your little jab there in an otherwise great reference
document posting seemed completely petty to me.  It makes me think
you stuck it in there as a little "See, I'm dissing them like I am
supposed to!" due to the fact that you posted the message to both
c.l.perl and c.l.python.

Whatever...  It just bugged me and really seems like that kind of
thing is the CAUSE of the Perl/Python geeky religious pettiness...in
a document showing how to move between both religious worlds no less!

OK, I am truly rambling now.



From srenner at lycosmail.com  Sat Aug 21 10:39:34 1999
From: srenner at lycosmail.com (srenner at lycosmail.com)
Date: Sat, 21 Aug 1999 14:39:34 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu>
Message-ID: <7pmdn5$lt0$1@nnrp1.deja.com>

The reactions Mr. Christiansen evokes may have more to do with his style
than his content. It strikes me as pompous, pretentious, hostile,
autodidactic, preening, vain and unpleasant. Perhaps this is not the
right newsgroup for him.

sr


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From htrd90 at zepler.org  Sun Aug 22 08:58:51 1999
From: htrd90 at zepler.org (Toby Dickenson)
Date: Sun, 22 Aug 1999 12:58:51 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <7ple6r$kf0@ds2.acs.ucalgary.ca>
Message-ID: <37bf6144.1666109@news.freeserve.net>

nascheme at ucalgary.ca (Neil Schemenauer) wrote:

>This is called tuple unpacking.  It is definitely cool.  In the
>newer versions of Python unpacking works on other sequences as
>well.  For example:
>
>	x, y, z = (1, 2, 3)
>	a, b, c = [1, 2, 3]

Cool, I never knew that worked with other sequences too!

I guess "_tuple_ unpacking" isn't the best name for this anymore ;-)

Thanks,


Toby Dickenson



From mwh21 at cam.ac.uk  Fri Aug 20 08:27:37 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Fri, 20 Aug 1999 13:27:37 +0100
Subject: "def" vs "sub" (was RE: More random python observations from a
 perl programmer)
In-Reply-To: <37bd4385@cs.colorado.edu>
References: <001601beeaef$e795f380$712d2399@tim> <37bd4385@cs.colorado.edu>
Message-ID: <Pine.LNX.4.10.9908201315440.6022-100000@atrus.jesus.cam.ac.uk>

On 20 Aug 1999, Tom Christiansen wrote:
> In comp.lang.python, 
>     "Tim Peters" <tim_one at email.msn.com> writes:
> :In Python and in Perl (at least in the Perl I last tried that under) that
> :creates cyclic garbage:  makefac's lexicals have to contain a binding for
> :fac, and fac has to point back to makefac's lexicals in order to resolve
> :up-level references.  And that's (unrecoverable trash) the reason Python
> :doesn't have lexical closures today.  
> 
> That's quite right.  Doesn't python have some similar issue related
> to the use of bound methods?

Not in the simple case, no.

I take it you're thinking a bound method has a reference to the instance
which has a reference to the method? That's not the case, for, if you have
a class like this:

class Account:
    def __init__(self,amount):
        self.amount = amount
    def withdraw(self,amount):
        self.amount = self.amount - amount

then executing 

W=Account(50).withdraw

gets you a bound method which has a im_self member which refers to an
Account object which has a __class__ member which refers to the Account
class which has a __dict__ member which contains a function "withdraw" (as
in plain Python function).

(with apologies to the authors of sicp for using that example yet again,
and to all concerned for the length of that previous sentence).

>>> Account(50).withdraw.im_self.__class__.__dict__["withdraw"]
<function withdraw at 87cd30>

No cycles, see?

internally-y'rs
Michael




From mnot at pobox.com  Tue Aug 10 09:21:46 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Tue, 10 Aug 1999 13:21:46 GMT
Subject: ANN: http header parsing
Message-ID: <04bb01bee333$4c1e3d40$0301a8c0@mnot.net>

I've been working for a little while on a module that will allow easy
parsing, manipulation and output of HTTP 1.1 headers (as defined in
RFC2616).

Although it's very alpha at this point (meaning: many bugs, interface will
probably change), I'd like to get some feedback and solicit ideas, criticism
and help from some other people. If it's done right, it could make life a
lot easier when writing CGI, clients, servers, etc.

As an example, here's some of the things you can do:

f = Headers()        # can instantiate with text or rfc822.Message instance
to parse
f.cache_control = {}
f.cache_control.max_age = 600
f.cache_control.max_age = f.cache_control.max_age - 60
f.etag = "fooboo"
f.etag.weak = 1
f.expires = time.time()
f           # output as below; can also use the putheaders() method to feed
into httplib
Cache-Control: max-age=540
ETag: W/"fooboo"
Expires: Tue, 04 Jan 1999 04:34:22 GMT


You can download from http://www.mnot.net/python/ ; you'll need both
httpheaders.py and httpheadertypes.py.

Please remember that these have NOT been extensively tested or documented,
and should not be used in a production system. I'm primarily interested in
feedback on the interface and the approach that I've taken.

Cheers,

--
Mark Nottingham
Melbourne, Australia
http://www.mnot.net/





From kistler at fnmail.com  Fri Aug 20 07:21:23 1999
From: kistler at fnmail.com (Per Kistler)
Date: Fri, 20 Aug 1999 13:21:23 +0200
Subject: Setting environment variables.
References: <37BD3429.7364B4F7@iinet.net.au>
Message-ID: <37BD3A33.6CAA145A@fnmail.com>

Oliver White wrote:
 eg, after
> running my python program as a child process, will the changes to the
> environment variable affect it's parent (the *term)? (suspecting no) Is
> there annother way to set the parent's?

I'm wondering, whether one could exec to a program which sets
ones env and then exec back to a shell, which would then have
the desired env. One would no more be in the original shell,
but the new one would have the desired env.
Maybe some guru could elaborate on that....

-Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From arcege at shore.net  Tue Aug 10 13:50:54 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 10 Aug 1999 17:50:54 GMT
Subject: What's a good way to get my IP from Python?
References: <37B04D5F.2F00EE6C@cisco.com>
Message-ID: <2EZr3.1342$EG4.232940@news.shore.net>

Aaron Rhodes <aarhodes at cisco.com> wrote:
: Hello,

: Does anyone know of a really good way to determine the
: IP address of the ethernet cards on a Linux system
: using Python only (i.e. no C code/parsing external binaries, etc.) ?

Hopefully this is portable, but my work PC is down and don't have
access to a Mac; but theoretically it should be.

  import socket
  def my_ipaddr(interface_hostname=None):
    # give the hostname of the interface you want the ipaddr of
    hostname = interface_hostname or socket.gethostname()
    s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
    s.bind(hostname, 0)
    ipaddr, port = s.getsockname()
    s.close()
    return ipaddr  # returns 'nnn.nnn.nnn.nnn' (StringType)

You aren't connecting to anything so this _should_ work even if you are
not on a network at the time (PPP), the interface just needs to be
plumbed and initialized.  The begged question would be: how do you get
the hostnames of multiple interfaces from within Python.  Anyone have a
good answer for that?

  -Arcege




From rthomas at uiuc.edu  Fri Aug 27 04:07:07 1999
From: rthomas at uiuc.edu (Ryan Thomas)
Date: Fri, 27 Aug 1999 03:07:07 -0500
Subject: make test: fcntl IOError:[Errno 22]
Message-ID: <37C64727.98860F3A@uiuc.edu>

Installing Python 1.5.2 on Mac OS X Server (Mach/BSD 4.4, descendant of
nextstep).

./configure       # went fine
make              # went fine except "make[1]: [add2lib] Error 1
(ignored)"
make install      # went fine (no errors)

Later ran into a problem with fcntl (using mailman).

So, I ran make test and get this:

test test_fcntl crashed -- exceptions.IOError : [Errno 22] Invalid
argument

I searched for fcntl issues and saw recent thread "Python 1.5.2 on HP-UX
10.20" which discusses the *exact same error*.  Coincidence?
Unfortunately, the fixes there don't seem to apply (see my FCNTL.py
appended and /tmp is lockable, I assume).  (?)

Advice?  This is Mac OS X Server (descendant of nextstep).

Thanks.
-Ryan


# Generated by h2py from /usr/include/fcntl.h

# Included from sys/fcntl.h

# Included from sys/types.h

# Included from sys/cdefs.h
def __P(protos): return protos

def __STRING(x): return #x

def __P(protos): return ()

def __STRING(x): return "x"

def __attribute__(x): return

def __COPYRIGHT(s): return __IDSTRING(copyright,s)

def __RCSID(s): return __IDSTRING(rcsid,s)

def __SCCSID(s): return __IDSTRING(sccsid,s)

def __PROJECT_VERSION(s): return __IDSTRING(project_version,s)


# Included from machine/types.h

# Included from machine/ansi.h

# Included from machine/endian.h
def major(x): return ((int32_t)(((u_int32_t)(x) >> 8) & 0xff))

def minor(x): return ((int32_t)((x) & 0xff))

NBBY = 8
FD_SETSIZE = 256
O_RDONLY = 0x0000
O_WRONLY = 0x0001
O_RDWR = 0x0002
O_ACCMODE = 0x0003
FREAD = 0x0001
FWRITE = 0x0002
O_NONBLOCK = 0x0004
O_APPEND = 0x0008
O_SHLOCK = 0x0010
O_EXLOCK = 0x0020
O_ASYNC = 0x0040
O_FSYNC = 0x0080
O_CREAT = 0x0200
O_TRUNC = 0x0400
O_EXCL = 0x0800
FMARK = 0x1000
FDEFER = 0x2000
FHASLOCK = 0x4000
O_NOCTTY = 0
def FFLAGS(oflags): return ((oflags) + 1)

def OFLAGS(fflags): return ((fflags) - 1)

FAPPEND = O_APPEND
FASYNC = O_ASYNC
FFSYNC = O_FSYNC
FNONBLOCK = O_NONBLOCK
FNDELAY = O_NONBLOCK
O_NDELAY = O_NONBLOCK
F_DUPFD = 0
F_GETFD = 1
F_SETFD = 2
F_GETFL = 3
F_SETFL = 4
F_GETOWN = 5
F_SETOWN = 6
F_GETLK = 7
F_SETLK = 8
F_SETLKW = 9
F_PREALLOCATE = 42
F_SETSIZE = 43
F_RDADVISE = 44
F_RDAHEAD = 45
F_READBOOTSTRAP = 46
F_WRITEBOOTSTRAP = 47
FD_CLOEXEC = 1
F_RDLCK = 1
F_UNLCK = 2
F_WRLCK = 3
F_WAIT = 0x010
F_FLOCK = 0x020
F_POSIX = 0x040
F_ALLOCATECONTIG = 0x00000002
F_ALLOCATEALL = 0x00000004
F_PEOFPOSMODE = 3
LOCK_SH = 0x01
LOCK_EX = 0x02
LOCK_NB = 0x04
LOCK_UN = 0x08
O_POPUP = 0x80000000
O_NO_MFS = 0x40000000
O_ALERT = 0x20000000




From claird at starbase.neosoft.com  Thu Aug 26 18:09:55 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Aug 1999 22:09:55 GMT
Subject: Metacomment on discussion forums (was: Python article on Builder.com)
References: <7DB247B399FED211A73F00600841DDC7099717@TAL_ENT>
Message-ID: <FE1E10BA8FBD1F8C.58F8A2D148BFEB3F.260D4DDD23232FCE@lp.airnews.net>

In article <7DB247B399FED211A73F00600841DDC7099717 at TAL_ENT>,
Tom Funk  <tdfunk at asd-web.com> wrote:
			.
			.
			.
>Everyone should comment.  I did...
			.
			.
			.
Is there a different way to approach those "technical
forums"?  I find them--I lump Builder.com in with
slashdot, ...--very unappealing.  Usenet feels like
home.  Those other places, in contrast, are noisy,
slow, and frustrating.  I'm open to suggestions on
whether there's a technical fix that'll change my
perception, or whether it's a deeper cultural gap that
I can't or won't bridge.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From I.Clarke at strs.co.uk  Wed Aug 11 07:18:17 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Wed, 11 Aug 1999 12:18:17 +0100
Subject: Why is CGI module so complicated?
References: <37AB04A8.C3403427@NOSPAM.strs.co.uk> <cWYr3.1329$EG4.226668@news.shore.net>
Message-ID: <37B15BF8.B7204705@strs.co.uk>

> : Can anyone explain what the point of the FieldStorage and
> : MiniFieldStorage objects are?

> Ian, please read my reply in the previous thread:
>   http://www.dejanews.com/getdoc.xp/AN=502365247

Well, unfortunately that link didn't work, but I managed to find the
comment I believe you are referring to, in which case I assume you were
indicating that the FieldStorage and MiniFieldStorage objects were used
for the case where a file upload is taking place.

I can't help but feel that this is not the best way.  How about go with
the method I described in my first post (where cgi fields are returned
as the types that they actually are, rather than being encased in
MiniFieldStorage objects).  Were more than one item is selected for a
given field a list could be returned, and for a file upload a tuple
could be returned containing the filename and the contents of the file -
or if that is not considered extendable enough, a CGIFileUpload object
could be returned containing the necessary fields.

This would mean that for the vast majority of CGI fields which are just
normal text, retrieval is very straight-forward and intuitive, just like
accessing strings stored in a dictionary.  More complexity is only
introduced when the user wants to do more complicated stuff (and even
then, it is pretty simple).  As mentioned in my previous post, it would
also allow the fields to be used as the substitutions for a
TemplateDocument directly, which can be quite a useful trick.

If I get some spare time today I will create a small module which will
provide this interface to CGI (although if anyone else wants to do it,
feel free!).  It would obviously be better if (assuming I can convince
others that this is a better way) this was integrated into the built-in
cgi module.

Ian.



From I.Clarke at strs.co.uk  Mon Aug 23 05:29:17 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Mon, 23 Aug 1999 10:29:17 +0100
Subject: Good riddance
References: <199908201554.LAA11430@python.org> <37BDAD0A.FAFC980A@lineone.net> <37bf4ba1@cs.colorado.edu>
Message-ID: <37C1146D.394B5E75@strs.co.uk>

In comp.lang.python Tom Christiansen wrote:

> But the invective has turned me from this project.  I wanted to learn
> Python, and I wanted to get Perl programmers to learn it.  I can see
> that the viciousness of assholes (not you) is the same here as elsewhere,

By "elsewhere" I assume you include C.L.P.L in which case the most
vicious 
of the assholes is none other than Tom Christiansen himself!

> and I'll return now to my previous condition and residence.  Thanks
> to the collective you for such an unpleasant experience.    It's not
> the language.  It's the assholes.

Quite right, I am sure Perl is a very nice language really.

Ian.



From mnot at pobox.com  Sun Aug 15 02:40:26 1999
From: mnot at pobox.com (Mark Nottingham)
Date: Sun, 15 Aug 1999 06:40:26 GMT
Subject: Python CGI Scripts
References: <Pine.LNX.4.04.9908150140180.584-100000@gnr.u2me3.com>
Message-ID: <008701bee6e9$0efea350$0301a8c0@mnot.net>

Well, there is the cgi.py module, which is basically form parsing, etc. For
HTML formatting, you can use HTMLgen, which I think is what you probably
have in mind. See the contributed modules section of www.python.org

I'd suggest considering using a template class as well; IMHO, mixing HTML
and a scripting language is a bad idea, from a code maint / readability
standpoint (whether it's HTML in CGI, like this, or something like ASP/PHP).

More generally, yes, a lot of people use Python for CGI. Between cgi,
htmlgen, htmllib, urllib and httplib, there's pretty much everything you
need.


Cheers,



> I'm just getting into python really, and being a perler for a couple of
years
> I've gotten it to do just about everything I've ever needed for SA'ing. As
I
> thumb through 'Programming Python' it looks like just about everything
that can
> be done with perl can be done with python.
>
> But... have a lot of people adopted python as their new cgi language? This
is
> what I do primarily lately and I found only one program in the book called
> 'cgimail.py'. As I read through the code I notice that the h1 tags, the
form,
> the header infomation, etc etc are all being 'print'ed manually. Is there
no
> nifty CGI.pm equivelent in python?





From ajung at sz-sb.de  Thu Aug 26 09:48:40 1999
From: ajung at sz-sb.de (Andreas Jung)
Date: Thu, 26 Aug 1999 13:48:40 GMT
Subject: Tree module ?
Message-ID: <Pine.GSO.4.10.9908261547330.13215-100000@bonnie.sz-sb.de>

I'm looking for a general tree module for building non-binary unbalanced
trees. 

Any hints or comments ? 

Cheers,
Andreas
                                _\|//_
                               (' O-O ')
------------------------------ooO-(_)-Ooo--------------------------------------
   Andreas Jung, Saarbr?cker Zeitung Verlag und Druckerei GmbH
   Saarbr?cker Daten-Innovations-Center
   Untert?rkheimerstra?e, D-66103 Saarbr?cken, Germany
   Phone: +49-(0)681-502-1528, Fax: +49-(0)681-502-1509
   Email: ajung at sz-sb.de (PGP key available)
-------------------------------------------------------------------------------




From philh at vision25.demon.co.uk  Tue Aug 31 18:52:47 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 31 Aug 99 22:52:47 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <935457373snz@vision25.demon.co.uk> <gr6gq7.jd.ln@bigbird.sw.asiainfo.com>
Message-ID: <7qmnt8$df$3@newnews.ilk.de>

In article <gr6gq7.jd.ln at bigbird.sw.asiainfo.com>
           markh at usai.asiainfo.com  writes:
> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> : In article <7ps4ma$n06$1 at wagner.wagner.home>
> :            vitus at wagner.rinet.ru "Victor Wagner" writes:
> 
> :> Really it is very simple to define parrot interpreter on tcl - just
> :> define commands such as window, columnLayout, rowLayout as tcl procedures 
> :> which substitute
> :> something by default.
> 
> : Would anyone want to do this in Tcl as opposed to Python? (Not
> : wishing to get into another language war :-)).
> 
> Interestingly, your parrot specifications follow Tcl's syntax,

Great minds think alike, eh? :-)

BTW, I might have a bare bones implementation of the tkinter
backend up by this weekend, if all goes well.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From tchrist at jhereg.perl.com  Fri Aug 20 17:02:33 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Fri, 20 Aug 1999 15:02:33 -0600
Subject: making python man pages looks hard 
In-Reply-To: Skip Montanaro's message of Fri, 20 Aug 1999 11:40:15 -0500 (CDT) .
             <14269.33828.537574.712197@dolphin.mojam.com> 
Message-ID: <199908202102.PAA32032@jhereg.perl.com>

I've contacted the maintainer who will get back to you on it.

--tom



From quinn at cruzeiro.ugcs.caltech.edu  Sat Aug 21 22:56:16 1999
From: quinn at cruzeiro.ugcs.caltech.edu (Quinn Dunkan)
Date: 22 Aug 1999 02:56:16 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au> <7p7h3p$jto$1@newshost.accu.uu.nl> <37bdceb0.797602261@news.triode.net.au>
Message-ID: <slrn7rupml.sjv.quinn@cruzeiro.ugcs.caltech.edu>

On Fri, 20 Aug 1999 22:43:30 GMT, John (Max) Skaller <skaller at maxtal.com.au>
wrote:
>On 15 Aug 1999 23:05:29 GMT, m.faassen at vet.uu.nl (Martijn Faassen) wrote:
>
>>John (Max) Skaller <skaller at maxtal.com.au> wrote:
>
>>I'm sure type inference can be done (at least in a limited fashion), but
>>it would be nice to have an intermediate dropoff point; a viperi or
>>viperc that would work with type annotations. 

This is sort of of off-topic, but I'm just making sure that you're aware of
Dylan.  Dylan is (to me) philosophically very similar to python, but designed
from the ground up to support type declarations, etc., so it can be a
high-performance compiled language.  There have been many suggestions for a
high-performance version of python, much as I love python's syntax, it might
be less hairy and cleaner to just help out the gwydion people (harlequin has a
high-performance dylan compiler written in dylan, but it's not free).

Not python but at least a sibling or cousin or something.



From donn at u.washington.edu  Fri Aug 27 13:42:37 1999
From: donn at u.washington.edu (Donn Cave)
Date: 27 Aug 1999 17:42:37 GMT
Subject: Why do we call python a scripting language?
References: <37c6bb11.340373891@midpoint> <37c66347.518331@news.netset.com>
Message-ID: <7q6imd$kuo$1@nntp6.u.washington.edu>

Quoth tdixon.no at spam.fwi.com (Timothy Dixon):
| On Fri, 27 Aug 1999 10:18:55 GMT, guy_oliver at yahoo.com wrote:
|> Why do we call python a scripting language?  
|>
|> For me, and for most of the people I try to tell about the benefits of
|> python, the title 'scripting language' implies that its not a real
|> language capable of real programs.  It implies to most of them that it
|> is good for simple sysadmin and maybe short cgi scripts.
|
| I put it in the class of scripting languages because it's embeddable
| (as I understand it; I'm actually just a newbie) in other applications
| where it can be used for scripting another app.

Yes!  This is the useful meaning for "scripting", as I see it!  The
definitions that try to distinguish between ``quick, inconsequential''
vs. ``serious, production'' etc. are hopelessly lame, regardless of
their distinguished sources, because it's a useless struggle to assign
importance this way.  A 3 line Bourne shell script can be mighty
important production software.

I think this fits in with the other contexts where people use the term,
too.  I'm mostly a UNIX geek, but on platforms like MacOS, BeOS, AmigaDOS,
(Windows?) etc. I gather ``scripting'' is very much tied to the context
of some application.

But in these contexts the scripting language isn't always embedded, there
may (should) be other ways for it to be integrated with its application.
And as an embedded application language frankly I think Python isn't really
at its best.  The plasticity of languages like Lisp and Tcl make them easier
to adapt in some ways.  Like the UNIX shell is adapted to its application
environment, UNIX commands.

| It's not *just* in that class, of course; polymorphism and multiple
| inheritance are great real-world characteristics. :-)

Fer sure.  My vote would be to leave ``scripting'' out of it, or use
the term only with explanation.  The references to rapid prototyping,
etc., are somewhat along the lines I think people really have in mind
here, and perhaps someone who's fond of the term can articulate a useful
meaning for it without making it sound unduly trivial.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From skaller at maxtal.com.au  Fri Aug 20 18:50:32 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Fri, 20 Aug 1999 22:50:32 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> 	<m3vham44g5.fsf@atrus.jesus.cam.ac.uk> 	<14257.54137.72005.705719@anthem.cnri.reston.va.us> <37b6f1c2.347828751@news.triode.net.au>  <37B82032.E7A11398@starmedia.net> <slrn7rpbk8.7pl.neelk@brick.cswv.com>
Message-ID: <37bddb50.800834108@news.triode.net.au>

On 20 Aug 1999 00:40:04 GMT, neelk at brick.cswv.com (Neel Krishnaswami) wrote:


> Cecil is a dynamic OO language
>that Craig Chambers has been using to work out techniques for wringing
>speed out of dynamic languages. The only danger is that it will
>inspire John to add predicate methods to Python. :)

	Hm, I have enough ideas already for things to add :-)

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From Sven.Drescher at dlr.de  Mon Aug  2 10:53:15 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Mon, 2 Aug 1999 16:53:15 +0200
Subject: Libraries needed, please help!!!
References: <7ns6o5$jo2@news.go.dlr.de><004e01beda98$d14f9390$f29b12c2@secret.pythonware.com><7o3moo$dse@news.go.dlr.de> <14245.40279.79936.219772@weyr.cnri.reston.va.us>
Message-ID: <7o4ak3$imi@news.go.dlr.de>

Thanks!

But there are still errors.

My programm is very little, so I think, there are no syntax errors in it:

#include "Python.h"
int main(int argc, char **argv)
{
  Py_Initialize();
  PyRun_SimpleString("print 'Hallo, embedded world!'");
return 1;
}

Ok. I compiled and linked my program with the command:

CC -I/path/to/include/ -L/path/to/lib/ -lpython1.5 -o gui gui.cxx

errors:
undefined symbols ... first referenced in 'my_object_file.o'

The undefined symbols Py_Initialize and PyRun_SimpleString are still the
error messages.
What did I wrong?
Can anybody send me the compiler-options for embedding needed object-files
or other flags? Has anyone written a makefile, to do these work?

Thanks a lot!
Sven





From andy at robanal.demon.co.uk  Tue Aug 10 15:53:53 1999
From: andy at robanal.demon.co.uk (Andy Robinson)
Date: Tue, 10 Aug 1999 19:53:53 GMT
Subject: Printing from python on Windows platform
References: <Pine.LNX.3.96.990731200754.5616A-100000@Johann>
Message-ID: <37b08348.5158036@news.demon.co.uk>

Johann Spies <jhspies at futurenet.co.za> wrote:

>I am a Linux user writing a program for somebody that will use it on a
>Windows platform.  In Linux I use the following to redirect printing to
>the printer:
>
>    drukker = sys.stdout
>    drukker = os.popen('lpr','w')
>    drukker.write(drukstuk)
>    sys.stdout.write(drukstuk)
>
>What would the windows equivalent be?
>
>Johann

It depends very much how your printer is set up.    First of all, it
matters what you want to send.  If you have a Postscript file and
Postscript printer, or only want to print text with line breaks, then
you can write directly to it.  Thus with a Postscript printer on the
port LPT3, the DOS command
	COPY MYFILE.PS LPT3:
will do the job.

With a network printer, it depends what OS you are on.  On NT, you can
install Microsoft TCPIP Printing Services (Control Panel - Network -
Services - Add - Microsoft TCPIP Printing Services), whcih gives you a
good LPR command.   Having done that, you can also create new ports
mapped to IP addresses, to which files can be written.

If you really want to use the Windows print queue management features,
you need to use the WritePrinter API call.  Instead of sending it
straight to the printer, the job goes to the NT server which manages
the printer, letting people cancel the job later.  But you need to
write C to handle that stuff.

Most Windows apps don't spool a pre-existing file; they create a
printer device context and draw on it, which works on all versions on
Windows.  If you are working in Pythonwin, I can send you a code
snippet which does this.

Let me know what you want to do - I've spent a lot of the last six
months getting Windows to print stuff securely in the middle of the
night!

- Andy










From kernr at mail.ncifcrf.gov  Fri Aug 13 18:50:09 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Fri, 13 Aug 1999 22:50:09 GMT
Subject: More Win32 Binaries of Python Extensions Available
Message-ID: <37b49faa.9013938@news.erols.com>

Hello guys and gals,

I've added some binaries of Python extension modules to my page.

The old ones:
  * cephes 1.15
  * Multipack 0.7
  * FFTW-NumPy 0.6
  * signaltools 0.5
  * wstrop

The new ones:
  * Molecular Modelling Toolkit 2.0b9
  * ScientificPython 2.0b1
  * sparsemodule 0.2
  * the readline module

Paul Sokolovsky had originally ported the readline module to Windows.
His binaries unfortunately don't work on NT.  Hopefully, the only
change required was to compile against MSVCRT.dll.  If you run NT,
please try it and tell me about it.

Have fun!

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From claird at starbase.neosoft.com  Sun Aug 22 14:28:56 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Aug 1999 18:28:56 GMT
Subject: [Off-topic] Sysadmins (was Re: can some one help a newbie?)
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <7pmv4j$gtt@dfw-ixnews14.ix.netcom.com> <935273107snz@vision25.demon.co.uk> <7pp2fk$7ul@dfw-ixnews9.ix.netcom.com>
Message-ID: <7E3E0D54A8D288F4.715E9D960110B591.7A1C6BCD2B7DD054@lp.airnews.net>

In article <7pp2fk$7ul at dfw-ixnews9.ix.netcom.com>,
Aahz Maruch <aahz at netcom.com> wrote:
>In article <935273107snz at vision25.demon.co.uk>,
>Phil Hunt <philh at vision25.demon.co.uk> wrote:
>>In article <7pmv4j$gtt at dfw-ixnews14.ix.netcom.com>
>>           aahz at netcom.com "Aahz Maruch" writes:
>>> In article <935227166snz at vision25.demon.co.uk>,
>>> Phil Hunt <philh at vision25.demon.co.uk> wrote:
>>>>In article <7pkke6$lut at dfw-ixnews6.ix.netcom.com>
>>>>           aahz at netcom.com "Aahz Maruch" writes:
>>>>>
>>>>> Overall, if your background is as a sysadmin using shell scripts, grep,
>>>>> awk, sed, and so on, you'll find Perl a bit easier.
>>>>
>>>>I've used all these, and find Python easier than Perl. Perl would
>>>>probably be easier than Python for 1-liners, if I could remember
>>>>the syntax.
>>> 
>>> Note that I'm specifically assuming a sysadmin with little actual
>>> programming experience.
>>
>>The idea that a company would employ someone, who can't really code,
>>as a sysadmin is IMO frightening. Unless they don't care whether
>>their computer system works, of course.
>
>I guess I'm missing something, because the prospect doesn't "frighten"
>me.  There are a lot of activities for which a little bit of programming
>goes a long way, but I've seen a lot of people capable of doing that who
>really are not programmers in any sense of the word that you and I would
>use.  That's precisely a large part of why Perl has been so successful;
>it lends itself readily to the kind of "cookbook" programming that
>sysadmins do.
			.
			.
			.
Right.  There's probably some deep way to understand
all this that I haven't yet grasped.  Most of what I
know is that I have several friends and acquaintances
who like being system administrators, and they're
good at it, but claim to want no part of programming
(or at least not on the job).  The closest I've come
to integrating this is to think of high-performance
mechanics, who *really* like tuning vehicles as they
should be, but have little interest in personally
driving or flying them.  Those are categories that
feel homologous to me.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From tchrist at mox.perl.com  Thu Aug 19 13:30:51 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 10:30:51 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <19990819174800.D5227@xs4all.nl>
Message-ID: <37bc313b@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Thomas Wouters <thomas at xs4all.nl> writes:
:It's an extension of Python's setup. Not everything is done to be immediately
:obvious, but everything is done to be _consistent_, and instantly obvious if
:you understand python. 

This is the goal.  It is not always reached.  There are plenty of 
oddities sticking out.

:&& and || are and and or.

I know that.  I asked why.

:Only if you aren't used to exceptions. Exceptions are really very simple and
:nice things. Especially back when they were just strings (before my time, but
:still ;-)

No, I'm sorry, the expressivity of "break LABEL" and "continue LABEL"
are much clearer than the cicumlocutions necessary for exceptions.
That's like saying you can get the same effect by adding a bunch of lines
of code.  That's not the same as built-in support, nor half so clear,
nor half so convenient.  Isn't python about rapid prototyping and all?
If it were about purity over convenience, I'm sure it would have been
a lot different -- and completely unused.

:There are 'raw' strings though, 'r"<string>"', that dont do backslashes.

Those aren't the book, and there are no manpages.  Therefore, 
in many ways, they don't really count.  I know it hurts to hear
this, and you're all going to jump on me, but please please think
about it for a bit before you do so.

:> GOTCHA: (high)
:>     Python has no manpages!  The horror!!!!!!!!!!!!!!!!!!!!!!
:>     ENODOC
:
:Damn, i hadn't even noticed that. Can you imagine ? I guess the '__doc__'
:attributes on objects are enough manpage for me. ;)

It's unconscionable.   Lack of tools and the tool-based 
approach strikes again.  See other postings.

:> 	>>> for c in ("weird", "bits"):
:> 	>>> for c in ("weird"):

:Wait, wait, you're doing two different things here. First off, you called
:'for' with different types of arguments, the first time with a tuple of
:strings, the second type with just a string. 

Oh, come now, you're telling me I'm expected to count the items
in the parens to know whether a split will happen.  Whatever. 
It's not obvious, so it's a real gotcha.  This is not consistent.

:But they are definately not the same :)

But they look the same, which is wicked.

:>>> import copy
:>>> print copy.__doc__

That's not a fricking manpage.  How can I run "apropos copy"
on it?  How do I do 

    grep copy /usr/python/man/*/*.*

or 

    find /usr/python/man -print | xargs grep copy /dev/null

or 

    troff -man /usr/python/man/man3/copy.3py

etc etc.  Tools, man.  Man tools.  Flexible interchangeable
parts.

:"Yes, the trailing comma counts" :)

Speaking of inconsistencies!

:
:> QUESTION:
:>     What is and what is not thread safe?
:
:Everything is thread safe. Or rather, there will at all times be only one
:thread executing the Python interpreter.

Huh, no threads in python?

--tom
-- 
"The past is the only dead thing that smells sweet."
				- Edward Thomas



From fredrik at pythonware.com  Thu Aug 26 06:20:25 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 26 Aug 1999 10:20:25 GMT
Subject: binary data
References: <37C5169B.824BF1F3@xsoft.co.at>
Message-ID: <002f01beefac$9dcd3f80$f29b12c2@secret.pythonware.com>

Gerhard W. Gruber <g.gruber at xsoft.co.at> wrote:
> How do I write binary data to a file?
> I'm using open() and write and I always get the error that n is an
> integer.

1. make sure to open the file in binary mode (the 'b' flag).
see:

    http://www.python.org/doc/current/lib/built-in-funcs.html

2. for different ways to convert integers to binary data, see:

    http://www.python.org/doc/current/lib/module-struct.html
    http://www.python.org/doc/current/lib/module-array.html

</F>




From fukuda at jb3.so-net.ne.jp  Sat Aug 28 14:34:57 1999
From: fukuda at jb3.so-net.ne.jp (Taka Fukuda)
Date: 29 Aug 1999 03:34:57 +0900
Subject: HyperCard Like Application
Message-ID: <86u2pj7nzi.fsf@goliath.nomad.com>

    Hello,

    I am looking for a Python script which is similar to
    HyperCard on MacOS.  Suggestions about a good start
    point would be also helpful for me.

    I love both Macintosh and Linux.  Usually, I access
    Linux box via VNC viewer on Mac.  This could be the best
    combo environment if you like to use Mac and Linux, and
    if you can afford to space for two desktop computers.

    However, I am now wondering if I can abandon MacOS.
    This is because VNC is a little sluggish on my old
    Quadra 700 and and because I'd like to replace my
    PowerBook 540 with a righter note book which Apple
    won't(can't?) make.  Actually, now I can do almost
    anything on Linux.  The only exception is HyperCard; I
    have created more than 10,000 cards and I cannot throw
    them away.  MetaCard may be a replace, but it cannot
    handle two byte code properly (and is expensive too).
    
    Thanks in advance for your help.
    
--  
--                                          Taka Fukuda
--			            fukuda at computer.org
    



From thomas at xs4all.nl  Tue Aug 24 09:49:56 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Tue, 24 Aug 1999 15:49:56 +0200
Subject: Necessity of ``pass''
In-Reply-To: <hhvha5tswc.fsf@forcix.roof.lan>
References: <hhemgu41z7.fsf@forcix.roof.lan> <slrn7s408d.gvn.wtanksle@dolphin.openprojects.net> <hhvha5tswc.fsf@forcix.roof.lan>
Message-ID: <19990824154956.J28043@xs4all.nl>

On Tue, Aug 24, 1999 at 11:44:19AM +0200, forcer wrote:
> > Another example of redundancy in Python -- and one with just as much
> > chance of being changed -- is the ':' symbol.  Take a look at the following:
> 
> Though the colon really improves readability while not getting in
> your way :)

Actually, no :) the colon may be three whole characters less than the 'pass'
keyword, but your argument for skipping 'pass' goes for the colon just as
much:

if spam #and eggs() or ham and timber['victim']:
	# your code here

You can't just comment out a non-working test in your if statement (like
above), you have to add the : to the right place, AND remember to remove it
when you finish the remaining tests.

So, lets remove neither ;-P

Annoying-ly-y'rs,
-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From Gary.Pennington at uk.sun.com  Tue Aug 17 11:25:54 1999
From: Gary.Pennington at uk.sun.com (Gary Pennington)
Date: Tue, 17 Aug 1999 16:25:54 +0100
Subject: Problem trying to spawn (fork) asynchronous process from cgi
Message-ID: <37B97F02.C5A27998@uk.sun.com>

Hi,

I'm trying to spawn(Win32 equivalent of fork and exec) a process from
python running under NT using CGI. I've got almost everything correct,
but one last thing is troubling me.

I invoke the script, my background process launches asynchronously(using
hte P_DETACH option to os.spawnve), I display my message in the browser
with successful notification - however the browser thinks that a
transfer is still running and thus continues displaying busy. If you
terminate the transfer, you get a transfer interrupted message in the
browser, other than this there is no impact on the back end server
process.

It looks ugly and is bugging me, so is there a way to stop this
behaviour? I suspected that it was caused by inherited open streams (eg
stdin etc...,) but I closed these down in my forked program (I don't
need them anyway) and that didn't change the problem. Interestingly, the
browser knows that the back-end proces is still running and stops the
transfer activity when the back end process is shut-down.

Thanks for any pointers.

Gary





From robin at jessikat.demon.co.uk  Mon Aug 30 14:16:47 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 30 Aug 1999 19:16:47 +0100
Subject: 'if foo' calls getattr?!?
References: <joe-300819991050234291@chinacat.salk.edu>
Message-ID: <6ocvdDAPqsy3EwxA@jessikat.demon.co.uk>

In article <joe-300819991050234291 at chinacat.salk.edu>, Joe Strout
<joe at strout.net> writes
>[[ This message was both posted and mailed: see
>   the "To," "Cc," and "Newsgroups" headers for details. ]]
>
>Doing some profiling on Graphite this weekend, we found a very curious
>thing: a gazillion calls to __getattr__ on one of our classes, where we
>expected to find none.  We eventually tracked it down to a seemingly
>innocent test, basically, "if foo: self.myfoo = foo".  The "if foo" was
>invoking foo.__getattr__ (which is very expensive).
>
>I can think of no reason why it should be this way, but it is.  Below is
>a very simple stand-alone demonstration.  It leads me to the absurd
>idiom of saying "if id(foo) != id(None)" when I really mean "if foo",
>just because using id() avoids the __getattr__ call.
>
>Can anybody explain why this should be?  Is it a bug in Python?  Is
>there some other way to check whether I have a reference to an object
>that doesn't involve a function call?
>
>Confused,
...
It seems that it tries to call foo.__nonzero__()  and then foo.__len__()
to test whether foo is true in some sense.

I think these are standard behaviours for class instances. I think you
should consider using if not foo is None: pass
-- 
Robin Becker



From rimbault at cybercable.fr  Tue Aug 24 17:33:32 1999
From: rimbault at cybercable.fr (Jean-Claude Rimbault)
Date: Tue, 24 Aug 1999 21:33:32 +0000
Subject: Pro Python remarks to math ed folks (pointers)
References: <37c2a51e.18050990@news.teleport.com> <Ibjw3.680$v6.61454@typhoon.nyroc.rr.com> <37c2b213.49798667@news.teleport.com>
Message-ID: <37C30FAC.C542DB6C@cybercable.fr>

Kirby Urner wrote:

> "Darrell" <news at dorb.com> wrote:
>
> >Installed your stuff and changed a few things to get it to work.
> >Had to add this method too Vector.
> >   def display(self):
> >       print self.data
>
> That's cool.
>
> I've just been going:
>
> from vector import Vector
> v = Vector(quad=(1,0,0,0))
> print v.data
> (1.0, 0.0, 0.0, 0.0)

You could also add a __repr__ method and __str__ method to Vector:

   def __str__(self):    # convert object to string (for printing)
      return `self.data`

   def __repr__(self):  # convert object to string
      return 'Vector'+`self.data`

Then you would be able to type:

   >>> v=Vector(quad=(1,0,0,0))
   >>> v
   Vector(1, 0, 0, 0)    # __repr__ method is called
  >>> print v
  (1, 0, 0, 0)                # __str__ method is called


Jean-Claude.




From billtut at microsoft.com  Thu Aug 12 05:32:53 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Thu, 12 Aug 1999 09:32:53 GMT
Subject: Using JPython and COM together.
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB91087@RED-MSG-50>

You can't use the normal win32com modules with JPython.
You'll need to use MS's COM support in Java. This means you'll need to use
the MS JVM as well.

Here are some URLs that help explain how to use COM from Java: 
(Big huge note: I've never done this, good luck!)

http://msdn.microsoft.com/library/periodic/period97/html/javacom.htm
http://msdn.microsoft.com/library/devprods/vs6/visualj/vjcore/vjovrbuildingi
mportingcomobjects.htm

Bill
I'm just a developer, not PR, and definately not a Java support engineer.


> From: Carlos Kyoku [mailto:carlos.ors at usa.net]
> 
> 
> Hi!
> 
> I'm trying to access COM from jpython but I don't know how. I 
> tried to access
> using win32com modules which belongs python but I can't neither.
> 
> How can I acces COM from jpython?
> 
>                                          A lot of thanks.
> 



From ovazquez at sprint.ca  Thu Aug 12 02:12:34 1999
From: ovazquez at sprint.ca (Orlando Vazquez)
Date: Wed, 11 Aug 1999 23:12:34 -0700
Subject: Python tkinter with Tcl/Tk 8.1?
References: <7osg3n$l41$1@news.vanderbilt.edu>
Message-ID: <37B265D2.8A5C962A@sprint.ca>

foo wrote:
> 
> I could not find any documentation on this in the FAQ or the tkinter pages,
> so I am asking here. I have had to move to Tcl/Tk 8.1 for another software
> package I am using on my computer. I have rebuilt Python 1.5.2 after getting
> the new Tcl/Tk installed and (no great surprise) tkinter does not seem to
> work (I can successfully import Tkinter from the python command line, but
> attempting to run Tkinter._test() crashes python due to access violations
> (reading wild pointers).
> 
> Is it possible to build the tkinter module successfully with Tcl/Tk 8.1?
> 
> Thanks,
> Jonathan

Thats interesting. I've also had problems with tcl/tk 8.1. trying to
load
any tcl/tk apps results in errors. for example:

scratchy:/usr/local/src/Python-1.5.2/Demo/tkinter/guido$ python hello.py
Traceback (innermost last):
  File "hello.py", line 17, in ?
    main()
  File "hello.py", line 7, in main
    root = Tk()
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in
__init__
    self.tk = _tkinter.create(screenName, baseName, className)
TclError: Can't find a usable init.tcl in the following directories: 
    



This probably means that Tcl wasn't installed properly.

scratchy:/usr/local/src/Python-1.5.2/Demo/tkinter/guido$ 

I'm not an X person, most of the time I'm in the console, but i'm the
kind
of guy that likes to have everything working even if i dont need it. :-)
anyone know that the heck is going on?



From news at dorb.com  Thu Aug 26 15:54:37 1999
From: news at dorb.com (Darrell)
Date: Thu, 26 Aug 1999 15:54:37 -0400
Subject: Fuzzy string matching?
References: <199908261746.TAA25440@dionysus.fw.cuci.nl>
Message-ID: <WXgx3.1646$v6.107045@typhoon.nyroc.rr.com>

Your distance thing is cool.
But check out soundex.
These are a bit redundent.
http://www.outfitters.com/genealogy/what-soundex.html
http://www.darken.com/soundex.htm
http://www.mayrand.org/soundex.html
--
--Darrell
Hans Nowak <hnowak at cuci.nl> wrote in message
news:199908261746.TAA25440 at dionysus.fw.cuci.nl...
> ...Anyone know of such a beast in Python?
>
> The background is this... I want to write a deck managing program for
> Magic the Gathering (or other CCGs or similar games). The program has
> a collection of cards it knows (ideally, all kinds of cards in
> existance). Now I want people to enter decks, by entering names of
> cards. However, since some card names are quite difficult or have
> funny characters in them, I expect that some of their input will not
> be recognized in the card database. In such cases, I would like to
> present the user with a list of card names which match the input
> closely, but not entirely... strings that 'almost match'. OK, so it's
> just a fuzzy string match. Is such a thing done in Python already? If
> not, would it be difficult to do? (I obviously don't how to do it
> myself otherwise I would have done so already... :-/  )
>
> TIA,
>
> --Hans Nowak (zephyrfalcon at hvision.nl)
> Homepage: http://fly.to/zephyrfalcon
> Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
> The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/





From danny_shevitz at gmo.com  Wed Aug 11 10:41:24 1999
From: danny_shevitz at gmo.com (danny_shevitz at gmo.com)
Date: Wed, 11 Aug 1999 14:41:24 GMT
Subject: help learning python source code
Message-ID: <7os22i$9s5$1@nnrp1.deja.com>

I've decided to learn something about the python source code. Does
anyone have any suggestion on the best way to learn the code? Is there
a road map somewhere of how all the pieces fit together?

Any suggestions would be appreciated.

thanks,
Danny


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From budd at csa.bu.edu  Wed Aug 18 23:53:29 1999
From: budd at csa.bu.edu (Phil Budne)
Date: 18 Aug 1999 23:53:29 -0400
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <000101bee869$9058c760$6e2d2399@tim>
Message-ID: <xomg11go286.fsf@csa.bu.edu>

Tim Peters  <tim_one at email.msn.com> wrote;

> While I'm tickled by the image, nothing is like SNOBOL4 -- not even its
> modern successor, Icon.  SNOBOL4 was unique, and is still the Best Language
> on Earth for writing string pattern-matching algorithms.

There are those of us who still use and develop SNOBOL4.

I've done a free port of the original Macro SNOBOL4, souped up with
some features of SPITBOL. It will run pretty much on anything with
pointers that are at least 32 bits long, and has a C compiler; Macro
SPITBOL is available for a wide variety of platforms from Catspaw Inc,
who also runs a SNOBOL4 mailing list.

For further information see;

	http://snobol4.com
	http://people.ne.mediaone.net/philbudne/snobol.html

For those who like SNOBOL4 but need a dose of syntactic sugar should
try Andrew Koenig's SNOCONE.

-phil



From fredp at multimania.com.nospam  Sat Aug 28 16:52:02 1999
From: fredp at multimania.com.nospam (Fred Pacquier)
Date: 28 Aug 1999 20:52:02 GMT
Subject: Why do we call python a scripting language?
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com> <37c726c4.2148316@news.netset.com>
Message-ID: <8E30E89A8PaCmAnRDLM@news.dial.oleane.com>

someone at someplace.com (Someone) said :

>>for the profane. The one app that 'did it' for me, curiously, is PySol.
>>It's a game, but a game engine, and a powerful one. And it shuffles
>>around a lot of graphics (pun intended), which may surprise people
>>biased by the 'scripting language' tag.
>
>This brings me to a new point.  since I first used python, I've been
>wondering about distribution of my final products.  In general, i dont
>want to distribute a full python distribution and all my source, as
>much from not wanting to burden the user with this task as for
>protecting my source.  I've played with freeze, but since I never
>really had anything to distribute, I've never really used it for
>anything except play things.  But from what I remember, there is
>nothing preventing a person from putting into a small package
>everything needed to distribute an app.
>But what I find most interesting is that I havent seen much evidence
>that any one uses it.  Is any one using freeze or some derivative or
>work alike?  Most things I find are written for users of python, and
>are just distributed as source.

That's also part of it, I guess. Say you Whip out a nice little game, say a 
200K archive (py code, bitmaps, docs), then what next ? You put it up on 
your Web page, with a note saying "Oh, by the way, you also need to 
download and install the X-Meg Python engine and the Y-Meg wxPython library 
to run this on" ?
Probably a put-off for many non-Python casual users. Microsoft can get away 
with it (think VB, MFC DLLs...), but for everyone else...
I've never even looked into freeze, but my understanding is that it's not 
trivial to set up, or am I wrong ?

>>Yet another selling point might be if IDLE or PythonWin or some other
>>seed came to evolve into a full-blown IDE up to par with Microsoft's or
>>Borland's, that might impress the developer population. After all, VC++
>>is done in MFC/C++ and Delphi in ObjectPascal, right ? But I seem to
>>recall this was all discussed to death some months ago... :-)
>I'm very impressed with what I am seeing on the python win 2 beta
>site.  its very impressive!  

Ah, yes. I visited there some time ago and though I needed to try that out, 
but I keep forgetting...

-- 
YAFAP : http://www.multimania.com/fredp/



From dwelton at linuxcare.com  Wed Aug 11 14:50:17 1999
From: dwelton at linuxcare.com (David N. Welton)
Date: 11 Aug 1999 11:50:17 -0700
Subject: What's a good way to get my IP from Python?
References: <37B04D5F.2F00EE6C@cisco.com> <14256.27817.224703.683095@buffalo.fnal.gov>
Message-ID: <87iu6m8846.fsf@linuxcare.com>

Charles G Waldman <cgw at fnal.gov> writes:

> How about this?  I don't know if it's "really good" but it works,
> and it shows you how to do IOCTL's from Python.  (Lately, folks on
> the list have been suggesting writing a new extension module
> whenever you need to do such a thing, but this is IMO unattractive
> from the point of view of easy redistribution of code).

I actually wrote a basic ifconfig module, although it's probably not
as thorough as it could be...

Personally, I find C more fun/easier to work with at this level.

Hopefully, I will be able to release my work soon, and then people can
vote with their feet.  I have what I need, so I will make sure it
keeps working and doing what I want.

Ciao,
-- 
David N. Welton, Web Engineer, Linuxcare, Inc.
415.354.4878 x241 tel, 415.701.7457 fax
dwelton at linuxcare.com, http://www.linuxcare.com/
Linuxcare. At the center of Linux.



From floods at netplus.net  Fri Aug 13 20:10:24 1999
From: floods at netplus.net (Luke Loeffler)
Date: Fri, 13 Aug 1999 19:10:24 -0500
Subject: take the 20 second lang. survey
Message-ID: <37B4B3F0.77E6C59A@netplus.net>

Take the 20-second language survey.  No personal information must be
entered.  This is only a survey steming from my personal interest--it is
not affiliated with any organization.

http://www.recordingwebsite.com/language.html






From joe at strout.net  Mon Aug 30 13:50:23 1999
From: joe at strout.net (Joe Strout)
Date: Mon, 30 Aug 1999 10:50:23 -0700
Subject: 'if foo' calls getattr?!?
Message-ID: <joe-300819991050234291@chinacat.salk.edu>

[[ This message was both posted and mailed: see
   the "To," "Cc," and "Newsgroups" headers for details. ]]

Doing some profiling on Graphite this weekend, we found a very curious
thing: a gazillion calls to __getattr__ on one of our classes, where we
expected to find none.  We eventually tracked it down to a seemingly
innocent test, basically, "if foo: self.myfoo = foo".  The "if foo" was
invoking foo.__getattr__ (which is very expensive).

I can think of no reason why it should be this way, but it is.  Below is
a very simple stand-alone demonstration.  It leads me to the absurd
idiom of saying "if id(foo) != id(None)" when I really mean "if foo",
just because using id() avoids the __getattr__ call.

Can anybody explain why this should be?  Is it a bug in Python?  Is
there some other way to check whether I have a reference to an object
that doesn't involve a function call?

Confused,
-- Joe


getattrcount = 0

class Foo:
   def __getattr__(self,attr):
      global getattrcount
      getattrcount = getattrcount+1
      return self.__dict__[attr]

def testA():
   global getattrcount
   getattrcount = 0
   foo = Foo()
   for i in range(42):
      if foo: pass
   print "calls due to 'if foo':", getattrcount

def testB():
   global getattrcount
   getattrcount = 0
   foo = Foo()
   for i in range(42):
      if id(foo) != id(None): pass
   print "calls when using id:", getattrcount
   

testA()     # results in 84 calls!
testB()     # results in 0 calls.

-- 
,------------------------------------------------------------------.
|    Joseph J. Strout           Biocomputing -- The Salk Institute |
|    joe at strout.net             http://www.strout.net              |
`------------------------------------------------------------------'
Check out the Mac Web Directory!    http://www.strout.net/macweb.cgi



From C.Kotsokalis at ece.ntua.gr  Wed Aug 25 18:46:34 1999
From: C.Kotsokalis at ece.ntua.gr (Constantinos A. Kotsokalis)
Date: 25 Aug 1999 22:46:34 GMT
Subject: Mirrors?!
References: <37C4569D.10267F7B@megsinet.net>
Message-ID: <slrn7s8sec.cn.C.Kotsokalis@ppp05.softlab.ece.ntua.gr>

On Wed, 25 Aug 1999 16:48:30 -0400, David Oppenheimer wrote:
>Does anyone know where I can download Python 1.52 from somewhere besides
>Python.org's site?  I have tried unsucessfully now to download it 3
>times...it downloads at .3k per second and fails after around
>1600K...arghhh!!!!

Our mirror at NTUA, Athens, Greece is accessible at:
ftp://ftp.ntua.gr/pub/lang/python

  Hope this helps,
    --C

-- 
Constantinos A. Kotsokalis || C.Kotsokalis at ece.ntua.gr
National Technical University of Athens - GREECE
Electrical and Computer Engineering Department
"Bus error -- driver executed."



From philh at vision25.demon.co.uk  Tue Aug 10 19:15:18 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 10 Aug 99 23:15:18 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au>
Message-ID: <934326918snz@vision25.demon.co.uk>

In article <37b075ae.2367293988 at news.triode.net.au>
           skaller at maxtal.com.au "John (Max" writes:
> This is a request for comments on Viper, Yet Another implementation
> of Python. I'm building Viperi, the interpreter,  and I'm interested in
> comments on three issues:
> 
>         1) compatability with Python
>         2) Extensions
>         3) Optimisation
> 
> Viperi is being built as a prelude to implementation of an optimising
> compiler, Viperc. Unlike previous attempts at a compiler, Viperc will do
> type inference, inlining, and special case analysis,
> in an attempt to optimise generated code.
> 
> The interpreter aims to be compatible with 'pure python':
> python not using external C modules, and not fiddling
> implementation details too much. Some of the fiddles
> will work, and some will not. 
> 
> Compatibility:
>         1) exceptions: both string and class exceptions are supported,
>         the class exceptions are client defined in exceptions.py
> 
>         2) rebinding module variables after importing is not permitted
>         3) None is a keyword.
>         4) range is a keyword
> 
>         5) What restrictions can you live with?
> 
> Extensions:
>         1) full slicing semantics
>         2) optional values in dictionaries (Defaults to None)
>         3) 'in' applies to dictionaries 'as if a sequence of keys were used'
>         4) lexically scoped functions (including lambdas)
>         5) optionally typed function parameters
> 
>         6) what do you want?

C-style comments

sorted dictionary loops, eg:

for k, v in dict.sortedValues():

where this is sorted by keys.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From nbecker at fred.net  Mon Aug  9 08:34:48 1999
From: nbecker at fred.net (nbecker at fred.net)
Date: 09 Aug 1999 08:34:48 -0400
Subject: ANN: PIL 1.0 final
References: <008e01bedfec$7ffe7110$f29b12c2@secret.pythonware.com> <37aadb42.108851580@news.u-psud.fr> <m37ln9i4or.fsf@schwinger.harvard.edu> <018001bee0e2$14502520$f29b12c2@secret.pythonware.com>
Message-ID: <x887ln5xhcn.fsf@rpppc2.hns.com>

You need to add -L/usr/X11R6/lib -lX11 to the final link, if you're
using Linux.



From billtut at microsoft.com  Fri Aug 13 17:00:20 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Fri, 13 Aug 1999 21:00:20 GMT
Subject: Does Python run under Windows 2000? (build 2031)
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB910A2@RED-MSG-50>

I've installed Python and the Win32 extensions without a hitch on 2092 and
everything seems to be in working order since PythonWin started up
successfully. 

Bill

> -----Original Message-----
> From: James E. Green [mailto:jgreen at RobotVision.com]
> Sent: Friday, August 13, 1999 5:20 AM
> To: python-list at cwi.nl
> Subject: Does Python run under Windows 2000? (build 2031)
> 
> 
> I'm running Windows 2000 RC1 (build 2031, the very latest).  
> Will Python
> run under this OS?  Only NT4 is mentioned on the Python web page.
> 
> Thanks, Jim Green
> 
> 
> 



From godzilla at netmeg.net  Sat Aug 21 23:23:10 1999
From: godzilla at netmeg.net (Les Schaffer)
Date: Sun, 22 Aug 1999 03:23:10 GMT
Subject: carpe linguam
References: <199908201554.LAA11430@python.org> <37BDAD0A.FAFC980A@lineone.net> <37bf4ba1@cs.colorado.edu>
Message-ID: <m2hflsxzm2.fsf@netmeg.net>

[courtesy cc'ed to author]

>>>>> ">" == Tom Christiansen <tchrist at mox.perl.com> writes:

    >> But the invective has turned me from this project.  I wanted to
    >> learn Python, and I wanted to get Perl programmers to learn it.

damn, the first set of interesting posts in a long time on this
newsgroup, and the instigator flees for cover.

    >> Thanks to the collective you for such an unpleasant experience.
    >> It's not the language.  It's the assholes.

hey, if you do post more of this comparison stuff, how bout at least
give a heads up over here, so some of us can stop by your place and
learn something.

les 'why use trojans when you can use sheiks' schaffer 



From m.faassen at vet.uu.nl  Thu Aug 19 03:51:12 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Aug 1999 07:51:12 GMT
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk> <19990816205639474.AAA315.413@max41127.izone.net.au> <lDvYnDAz1Hu3EwB6@jessikat.demon.co.uk> <37BB2E1D.C0733F85@boeing.com>
Message-ID: <7pgd1g$hj6$1@newshost.accu.uu.nl>

Howard S.Modell <howard.s.modell at boeing.com> wrote:
[snip[
> Perl? Python?  If we all programmed in SNOBOL4 the way g-d intended, 
> we wouldn't have to worry about debates like that!

Funny, I read that 'the way g-d indented'. :)

Regards,

Martijn




From abigail at delanet.com  Sun Aug 15 06:44:37 1999
From: abigail at delanet.com (Abigail)
Date: 15 Aug 1999 05:44:37 -0500
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cf70@cs.colorado.edu>
Message-ID: <slrn7rd6ik.a5.abigail@alexandra.delanet.com>

John W. Stevens (jstevens at basho.fc.hp.com) wrote on MMCLXXIV September
MCMXCIII in <URL:news:37b4cf70 at cs.colorado.edu>:
'' 
'' But if you do like OO, don't Perl.

I actually agree with that. But then, for the same reason, I wouldn't
use Python either. You see, Perl's OO model was taken from Python.
I don't like Perl's OO model, because 1) it doesn't give me what I want
from an OO model and 2) it doesn't give me what the rest of Perl does.

It's very unlikely I would turn to Python for its OO.



Abigail
-- 
sub camel (^#87=i at J&&&#]u'^^s]#'#={123{#}7890t[0.9]9 at +*`"'***}A&&&}n2o}00}t324i;
h[{e **###{r{+P={**{e^^^#'#i@{r'^=^{l+{#}H***i[0.9]&@a5`"':&^;&^,*&^$43##@@####;
c}^^^&&&k}&&&}#=e*****[]}'r####'`=437*{#};::'1[0.9]2 at 43`"'*#==[[.{{],,,1278@#@);
print+((($llama=prototype'camel')=~y|+{#}$=^*&[0-9]i@:;`"',.| |d)&&$llama."\n");


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From mkrishnam at my-deja.com  Wed Aug  4 19:28:59 1999
From: mkrishnam at my-deja.com (mkrishnam at my-deja.com)
Date: Wed, 04 Aug 1999 23:28:59 GMT
Subject: Problems with sockets on Win NT
Message-ID: <7oaibf$763$1@nnrp1.deja.com>

Hi,

I'm trying to exchange datagram (UDP) packets
between a unix machine and a Win NT machine.

The UDP packets sent from Win NT machine are
received properly on unix machine. However UDP
packets sent from unix machine is not at all
received on WinNT machine.

I'm using recvfrom method to receive the packets.

The code I'm using is as below.

# Code on Win NT machine

from socket import *

my_sock = socket(AF_INET, SOCK_DGRAM)
my_sock.bind('localhost', 6123)

while(1):
        test = my_sock.recv(1024)
        print "got it!"

# Code on Unix machine

from socket import *

my_sock = socket(AF_INET, SOCK_DGRAM)
my_sock.bind('localhost', 6440)

dst_addr = 'leto',6123

while (1):
        print my_sock.sendto("Hello", dst_addr)

A help is kindly appreciated.

Thanks
- Murali


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kistler at fnmail.com  Thu Aug 19 20:30:23 1999
From: kistler at fnmail.com (Per Kistler)
Date: Fri, 20 Aug 1999 02:30:23 +0200
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
Message-ID: <37BCA19F.F0EBE23E@fnmail.com>

Tom Christiansen wrote:

> This article was posted to the Perl newsgroup, but I have duplicated
> the posting here without crossposting to avoid flame wars.

Thanks for this posting. Since I'm a pragmatic programmer, I'd like
any information possible. Such infos translate to money, while it 
is more difficult to convert the other type of postings
here into some visible added value.

We could use the same energie which goes into flaming for 
creating new design patterns or sophisticated applications or
whatever makes life interesting. All the scripting languages
have long todo lists (, not tupples).

Once they prohibited me to use more than one language for 
a project; I had to stay with perl and even without OO.
The more languages, the better for fighting alzheimer:-)

Cheers, Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From claird at starbase.neosoft.com  Mon Aug 23 00:35:35 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Aug 1999 04:35:35 GMT
Subject: 16-bit Python -- does it exist?
References: <MPG.122214458f7ab0d2989685@news.mop.no> <37B81212.733CDD9B@interet.com> <MPG.1224b6155e949206989686@news.mop.no>
Message-ID: <AE8E51894623103B.DF110B5935EE4519.C605411C103710B2@lp.airnews.net>

In article <MPG.1224b6155e949206989686 at news.mop.no>,
Alexander Staubo <earlybird at mop.no> wrote:
>In article <37B81212.733CDD9B at interet.com>, jim at interet.com says...
>> Alexander Staubo wrote:
>> > 
>> > Subject says all. A friend (really!) would like to run Python on an old
>> > 80286. Possible?
>> 
>> Python for 32-bit DOS and for Windows 3.1 are on the
>> download page www.python.org/download.
>
>Read the subject again. 32-bit Python won't run on a 286.
			.
			.
			.
The URL is a useful one, though.  You'll find lots
of variations of interest (memory extenders?  OS/2?).
I'm running a version 1.0.1 port <URL:http://
www.python.org/ftp/python/pc/16python.exe.gz> on a
few 8086 and 8088 space heaters I maintain.  They
become very handy calculators.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From tim_one at email.msn.com  Tue Aug 17 21:20:23 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 21:20:23 -0400
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <14265.58497.578663.958576@buffalo.fnal.gov>
Message-ID: <000701bee917$d8161cc0$eaa22299@tim>

[Charles G Waldman, making heroic efforts to test the quality of
 random-permutation algorithms]
> ...
> Another, more straightforward way to test sampling strategies is just
> to keep track of how many times each permutation comes up.  However,
> since N! gets so large, it takes too long to build up decent
> statistics this way.

I'll save you a near-infinity of work, then <wink>:  Python's Wichmann-Hill
uniform generator has a period of around 1e12 to 1e13.  Since the point at
which you begin in the period completely determines the permutation that
will be generated, you can't get more than 1e13 distinct permutations out of
it.  So a sequence of length 16 is the longest for which it could even
conceivably produce all permutations; if you use it to, e.g., shuffle a deck
of cards, the number of permutations any algorithm based on WH can get back
is a vanishingly small fraction of the possibilities (since 52! ~= 1e66; we
can only get back at most 1 of each 1e66/1e13 = 1e53 possibilities).

So, in a real sense, even the best possible algorithm based on
random.random() is terrible, incapable of generating "almost all" of the
space for a measly deck of cards.

It's surprising that so few people get disturbed by this; but it's even more
surprising that some people do <wink>.

the-deck-is-always-rigged-in-subtle-ways-ly y'rs  - tim





From abigail at delanet.com  Sun Aug 15 06:23:20 1999
From: abigail at delanet.com (Abigail)
Date: 15 Aug 1999 05:23:20 -0500
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7rb1nc.cf9.jstevens@bamboo.verinet.com>
Message-ID: <slrn7rd5al.a5.abigail@alexandra.delanet.com>

John Stevens (jstevens at bamboo.verinet.com) wrote on MMCLXXIV September
MCMXCIII in <URL:news:slrn7rb1nc.cf9.jstevens at bamboo.verinet.com>:
$$ On 14 Aug 1999 04:08:05 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
$$ >That could simply have been a reference. Or a symbolic reference.
$$ >
$$ >What is fundamental is that a @ tacked on the front indicates that it is an
$$ >array.
$$ 
$$ What is so amusing about that, is that you can say that with a straight
$$ face!

It's nice to see you have fun about nothing. 

$$ >So given @$fred, even with no knowledge of what that exactly means
$$ >you should be able to tell that it is somehow treating $fred as an array.
$$ 
$$ No, what any reasonable person would do would be to grab for his
$$ Perl book. . .

Perhaps the first and the second time he encounters it. If he needs it
a third time, he sucks as a programmer. How often have you looked this
up? Just what I said.

$$ >>Yes. . . is it a hash, or a scalar?  If it is a scalar, why
$$ >>is it called dict?  If it is a hash, then why is it prefixed
$$ >>by $?  If this is a reference instead of a scalar, then why
$$ >>doesn't it have it's own special prefix character.  ;->
$$ >
$$ >It's a scalar. It is named dict because TomC called it that.
$$ 
$$ Yes.  My point exactly.
$$ 
$$ >It is
$$ >also named that since it is a reference to a hash. I use code like this
$$ >in C quite a bit :
$$ 
$$ A reference to a hash. . . and yet TC claims that Perl is open to
$$ non-computer scientists.

Yep. Only computer scientists use pronouns. It's too difficult for the
rest of the people. Pronouns were invented by Turing, in the early 50s.

$$ Doesn't *ANYBODY* else see the irony in that?

No. What's the reference phobia? Does it give you spots?

$$ >If you know what it means then why do you continually get it wrong
$$ >throughout this thread?
$$ 
$$ I don't suppose that you realize that getting wrong simply
$$ proves (and illustrates) my point?
$$ 
$$ I learned it.  I used it.  I haven't written a new Perl program
$$ in three months.
$$ 
$$ I come back to it, I get it wrong. . . do you see, yet,

Yes, we see. You suck as a programmer.

$$ >Here is some code from Damian Conway from the 'Impythonating PERL' thread
$$ >in march.
$$ >
$$ >package impythonate;
$$ >use Text::Tabs;
$$ >my ($active, @bracket) = (0, ('{', ';', '}') );
$$ >sub import
$$ >{
$$  ^ Look closely. . . see that curly brace?
$$ 
$$ >And here is his sample code that is now valid perl (although anyone who uses
$$ >it for real code should be killed) :
$$ 
$$ To late.  You already used a curly brace.  Disproving your point,
$$ in case you hadn't realized it.


It looks like you have a problem following a linear argument.



Abigail
-- 
sub f{sprintf$_[0],$_[1],$_[2]}print f('%c%s',74,f('%c%s',117,f('%c%s',115,f(
'%c%s',116,f('%c%s',32,f('%c%s',97,f('%c%s',0x6e,f('%c%s',111,f('%c%s',116,f(
'%c%s',104,f('%c%s',0x65,f('%c%s',114,f('%c%s',32,f('%c%s',80,f('%c%s',101,f(
'%c%s',114,f('%c%s',0x6c,f('%c%s',32,f('%c%s',0x48,f('%c%s',97,f('%c%s',99,f(
'%c%s',107,f('%c%s',101,f('%c%s',114,f('%c%s',10,)))))))))))))))))))))))))


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From nika at kassube.de  Sat Aug 21 13:33:31 1999
From: nika at kassube.de (Nils Kassube)
Date: Sat, 21 Aug 1999 19:33:31 +0200
Subject: MacPython: Tkinter vs. CLI
Message-ID: <7pmnpj$s19$1@photon.on-luebeck.de>

Guten Abend.

I'm learning Python using the excellent "Learning Python"
book by Mark Lutz and David Ascher. Unfortunately, it was
only necessary to read a few pages to hit something I'd
call strange behaviour in the Mac port:

--cut--
from Tkinter import *
w = Button(text="Hallo", command='exit')
w.pack()
w.mainloop()
--cut--

If I drop a file containing these lines on the Python
interpreter, it works as expected. But if I type it
in Python's CLI it opens an inactive and empty window
after line 2 and then the CLI stops reacting to keyboard
input. You can still quit the Python application via the
"Quit" menu item after clicking in the Tk window, then
it's possible to use File/Quit to exit the interpreter.

Useful information:

Python 1.5.2b1 (#47, Jan 13 1999, 15:14:59)  [CW PPC w/GUSI
w/MSL]
Mac OS 8.6 (German), 25 MB allocated to Python

Cheers,
Nils
--
L?beck, Germany - The Marzipan City
http://www.luebeck.de
 



From herzog at online.de  Thu Aug 19 14:58:03 1999
From: herzog at online.de (Bernhard Herzog)
Date: 19 Aug 1999 20:58:03 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <19990819174800.D5227@xs4all.nl>
Message-ID: <m39077simc.fsf@greebo.nodomain.de>

Thomas Wouters <thomas at xs4all.nl> writes:

> On Thu, Aug 19, 1999 at 08:22:49AM -0700, Tom Christiansen wrote:

> > DISSIMILARITY:
> >     There are no private variables in a module or a class.
> >     They can always be fully-qualified and accessed from
> >     without.  Perl can do this with file-scoped lexicals,
> >     but still can't do so with data members (modulo obscure
> >     tricks such as using a non-hash for an object).
> 
> There is some namemangling though, for datamembers/functions that start with
> a single _, for hiding/overriding purposes. (those mmebers get the classname
> encoded in, so it's dependant on the class it was called on.)

For class members it's a leading double underscore.

The single leading underscore only affects module globals in that they
are *not* imported by 'from module import *'


> > GOTCHA: (low)
> >     You have insert and append methods for lists, but only
> >     a del function.  Why is it a function not a method, and

del is a statement, not a function. 

> > GOTCHA: (high)
> >     Because you can't use readline() to get a number, people seem to enjoy
> >     calling eval() just to get a string turned into a number:
> > 	import sys
> > 	str = sys.stdin.readline()
> > 	num = eval(x)
> >     This is scary.  Even scarier is the propensity for calling input(),
> >     which auto-eval()s its input to make sure it's the right "type".
> >     (Funny that a soi-disant "typeless language" should be so danged
> >     picky about this.)  That means you see people write:
> > 	num = input("Pick a number? ")
> >     But the person can supply as an answer 0x23 or 2+9 or pow(2,4)
> >     or os.system("rm -rf *").  I am stunned.
> 
> Try eval(number, {}, {}). No sys or os module to call, no other variables
> what so ever. 

Passing empty dicts is not enough. You still have access to all builtin
functions:

>>> s = '''eval(compile("import os; os.system('pwd')", "", "exec"))'''
>>> eval(s, {}, {})
/home/herzog

You can disable the builtins by defining a '__builtins__' item in the
globals dict:

>>> eval(s, {'__builtins__':{}}, {})
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<string>", line 0, in ?
NameError: eval
>>> 

See the rexec module.


The safest way to read a number interactively is probably to use
raw_input and to convert the string explicitly to an int or float with
int() or float() respectively. You loose the automatic recognition of
ints and floats and syntaxes for octal or hexadecimal numbers, though.

A standard function eval_literal() that works like eval() but only
accepts literals would be convenient for this. E.g.

eval_literal('1.0')		-> float 1.0
eval_literal('0x10')		-> int 16
eval_literal('(10, "hallo")')	-> tuple with int a string
eval_literal('sin(0.5)')	-> raise exception


-- 
Bernhard Herzog	  | Sketch, a drawing program for Unix
herzog at online.de  | http://www.online.de/home/sketch/



From never at mail.matav.hu  Sat Aug  7 12:00:42 1999
From: never at mail.matav.hu (Hever Zsolt)
Date: Sat, 7 Aug 1999 18:00:42 +0200
Subject: Help. Tracking ButtonRelease events
Message-ID: <199908071600.MAA10049@python.org>

This is not  a direct answer, but hopefully it is a help.
I am working on a bit similar program and I have put out some code.
http://www.nexus.hu/hever/prog/vu/visualut.py <=main program
http://www.nexus.hu/hever/prog/vu/frameobj.py <= module
http://www.nexus.hu/hever/prog/vu/vu.gif    <= help, a picture of the
program
It grew out from the same script(s) you mentioned, there are same problems
you have now that I had to solve somehow.
In the frameobj.py module there is a "class Connecting_Objects:" which does
that. I maintain an ordered taglist for additional information.

Zsolt

----------
> From: Philip Derbeko <sphilip at techst02.technion.ac.il>
> To: python-list at python.org
> Subject: Help. Tracking ButtonRelease events
> Date: 1999. augusztus 7. 14:09
> 
> Hello.
> 
> I'm writing an application part of which is a GUI to draw graphs.
> What i want is create verticies and edges between them,
> i create vertex by clicking on Button 1 ( taken from Demo/Tkinter/matt
> scripts )
> and edge created by clicking on Button 3 and dragging the mouse to the
> destination vertex.
> In order to do it every vertex  binded events ButtonPress-3 and
> ButtonRelease-3 ,
> the problem i run into is that when i press button 3 and drag it to the
> destination vertex,
> source vertex gets the ButtonRelease-3 event and not destination.
> 
> Am i missing something and how can it be solved ?
> 
> Thanks in advance
> Philip
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list



From tchrist at mox.perl.com  Thu Aug 19 22:44:43 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 19:44:43 -0700
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <c75yafced1p.fsf@pc142.cosc.canterbury.ac.nz> <37bb6923@cs.colorado.edu> <c75g11f1doq.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <37bcb30b@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Timothy R Evans <tre17 at pc142.cosc.canterbury.ac.nz> writes:
:x = 5
:def foo():
:    print x # works, prints 5
:def bar():
:    x = 6 # works, but doesn't change the global
:def wibble():
:    print x
:    x = 6 # this fails with NameError
:
:Seems like a good behaviour as it stops you doing dumb things.

Perhaps it could give a better error message, pretty please?  Sometimes
driving Python feels  like driving Ken Thompson's car.  It just keeps saying
"?" and you're expected to know what that means.  :-(


--tom

PS: From an old joke about Ken's treatment of error messages in ed.
-- 
    If I allowed "next $label" then I'd also have to allow "goto $label",
    and I don't think you really want that...  :-) [now works in perl5!]
            --Larry Wall in <1991Mar11.230002.27271 at jpl-devvax.jpl.nasa.gov>



From fdrake at cnri.reston.va.us  Tue Aug  3 10:52:26 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Tue, 3 Aug 1999 14:52:26 GMT
Subject: Libraries needed, please help!!!
In-Reply-To: <7o5vvh$lai@news.go.dlr.de>
References: <7ns6o5$jo2@news.go.dlr.de>
	<004e01beda98$d14f9390$f29b12c2@secret.pythonware.com>
	<7o3moo$dse@news.go.dlr.de>
	<14245.40279.79936.219772@weyr.cnri.reston.va.us>
	<7o4ak3$imi@news.go.dlr.de>
	<37A5B809.A29DDFF5@GermanLloyd.org>
	<7o5vvh$lai@news.go.dlr.de>
Message-ID: <14247.554.177824.770257@weyr.cnri.reston.va.us>

Sven Drescher writes:
 > > Order matters? Try:
 > >
 > > CC -I/path/to/include/ -L/path/to/lib/ -o gui gui.cxx -lpython1.5
 > >
 > > Good luck
 > >
 > > Berthold
 > > --
 > > email: hoel at GermanLloyd.org
 > >    )
 > >   (
 > > C[_]  These opinions might be mine, but never those of my employer.
 > 
 > I tried, but there are more undefined symbols yet. All symbols are first
 > referenced in libpython1.5.a(socketmodule.o)!

  Are the missing symbols socket-related (gethostbyname, etc.)?  If
you're doing all this on Solaris, you need to link with "-lsocket
-lnsl" to get all the symbols.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From phd at emerald.netskate.ru  Mon Aug 23 04:20:59 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Mon, 23 Aug 1999 12:20:59 +0400 (MSD)
Subject: Date Calculations
In-Reply-To: <37BF5D1A.9865CDF@fnmail.com>
Message-ID: <Pine.LNX.4.04.9908231218230.7587-100000@emerald.netskate.ru>

On Sun, 22 Aug 1999, Per Kistler wrote:
> I'm making some Date Calculaton class with things like
> deltaDays, addDays, daysInYear and so on, above the calendar 
> and time classes. After some 300 lines I thought, I better
> ask, whether that has already been done by someone else.
> How is this?

   I have a small module for date/time arithmetic. It is not OO - just a
set of functions. Look for opdate.py in m_lib.tgz on my homepage and
mirrors:

http://sun.med.ru/~phd/Software/Python/
http://members.xoom.com/phd2/Software/Python/
http://www.fortunecity.com/skyscraper/unix/797/Software/Python/

> Regards, Per.
> -- 
> Per Kistler kistler at fnmail.com / kistler at gmx.net

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From dave at see.sig.for.address  Fri Aug 20 13:20:55 1999
From: dave at see.sig.for.address (Dave)
Date: Fri, 20 Aug 1999 17:20:55 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <37bd557b.6915234@read.news.global.net.uk>

On 19 Aug 1999 08:22:49 -0700, Tom Christiansen <tchrist at mox.perl.com>
wrote:

>Per your collective requests, here is one of the documents
>with my collective observations from learning Python.
>
>This article was posted to the Perl newsgroup, but I have duplicated
>the posting here without crossposting to avoid flame wars.
>
>--tom
[long and interesting (if occasionally biased and out of date)
comparison snipped]

An interesting post (along with the previous one), and one I would
have found useful a couple of years ago when I moved from perl to
python. Some of the gotchas are ones that I fell prey to before I got
into the python way of thinking.

It would be an interesting exercise to write a companion document for
someone going from python to perl, and would be of some use to the
(admitedly far fewer) programmers making that transition. Any takers?

	Dave K
N.B. please dont cc any replies to me, since I read the NG regularly
--------------------------------------------------
All great ideas start as heresy and end as dogma.

dkirby@   <-figure this out, spambots!
bigfoot.   My opinions are my own,
com        but I'm willing to share.



From knotwell at knotwell.ix.netcom.com  Thu Aug 26 16:00:11 1999
From: knotwell at knotwell.ix.netcom.com (User Knotwell)
Date: 26 Aug 1999 13:00:11 -0700
Subject: What is a troll by the way? [offtopic]
References: <zelv3.13444$x04.816752@typ11.nn.bcandid.com> <37bdeb1b.31915768@news.erols.com> <37BDF3A7.75FEA04F@vision.arc.nasa.gov> <Z1xv3.11$EN1.7779@news.shore.net>
Message-ID: <86wvui71o4.fsf@knotwell.ix.netcom.com>

"Michael P. Reilly" <arcege at shore.net> writes:
> Anyone remember BIFF?

BIFF'S A C00L D00D!!!  
PSUVM ROX.  
WINDOZE SUX.  

What was PSUVM anyway?

>   -Arcege

I'll see your BIFF and raise you one better.  Does anyone remember alt.flame's
Asmodeus "Naked Under Leather?"

--Brad (hoping I'm not drooling)





From thomas.heller at ion-tof.com  Wed Aug 25 10:53:11 1999
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Wed, 25 Aug 1999 16:53:11 +0200
Subject: Making A Standalone Executable Under Windows NT/95/98
References: <hUQw3.117$Kl.24641667@speed.city-net.com>
Message-ID: <007e01beef09$8d424ff0$4500a8c0@thomasnotebook>

[Making A Standalone Executable Under Windows NT/95/98]

> Is this possible?  If so, How?

Look at Gordon McMillan's win32 installer
    http://starship.python.net/crew/gmcm/install.html.
Additional features are included in an enhanced version of this tool at
    http://starship.skyport.net/crew/theller/installer/.


Thomas Heller





From claird at starbase.neosoft.com  Mon Aug 23 00:19:09 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Aug 1999 04:19:09 GMT
Subject: Help: sendmail under MS NT
References: <7p1pma$ak6$1@nnrp1.deja.com> <37b4ca58.2777926@nntp.best.ix.netcom.com> <XM8t3.62641$jl.39581340@newscontent-01.sprint.ca>
Message-ID: <03D88C5655653B10.023A672131D906A8.E4D77C3EC5616AFD@lp.airnews.net>

In article <XM8t3.62641$jl.39581340 at newscontent-01.sprint.ca>,
Bruce Dodson <bruce_dodson at bigfoot.com> wrote:
>Check the topic indexes at C/C++ User's Journal and Dr. Dobb's Journal web
>sites.  Sometime between mid 1996 and early 1998, there was an article
>featuring a sendmail DLL for Win32, which hid the messy internals of MAPI
>and left you with something basically as simple as sendmail.  It would be a
>trivial matter to encapsulate this and the Unix sendmail in a portable
>least-common-denominator wrapper.
>

It's often even easier than this.  While the original
questioner asked about the "sendmail utility on NT"
(which *is* available, by the way, at a price), I
speculate it's likely he'd been happy simply pointing
a very simple network client toward the nearest open-
to-his-application SMTP relay.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From che at debian.org  Mon Aug  9 17:42:44 1999
From: che at debian.org (Ben Gertzfield)
Date: 09 Aug 1999 14:42:44 -0700
Subject: Packing and unpacking IP addresses into 32-bit integers
References: <ytt1zdcafmg.fsf@gilgamesh.cse.ucsc.edu> <Pine.A32.3.90.990809163915.17213C-100000@elvis.med.Virginia.EDU>
Message-ID: <yttyafk8wbv.fsf@gilgamesh.cse.ucsc.edu>

Thanks. Here's what I came up with finally. I decided to avoid use
of apply() for clarity, as these are dense enough as it is. :)

import string, struct

def inet_aton(packme):
    """Given a IP address as a string of the form 123.45.67.89, pack
       it into the 32-bit binary representation (struct in_addr) used
       in C and other languages."""
    
    a, b, c, d = map(string.atoi, string.split(packme, '.'))
    return struct.pack('BBBB', a, b, c, d)

def inet_ntoa(packed):
    """Unpacks a 32-bit binary representation of an IP address into a
       string and return it in dotted-quad (123.45.67.89) format."""
       
    quads = map(str, struct.unpack('BBBB', packed))
    return string.join(quads, '.')

I'm not sure if it's overly-compact, or if I ought to do things with
binary arithmetic for clarity. Anyway, it seems to work. Thanks for
the tip!

Ben

-- 
Brought to you by the letters O and Q and the number 5.
"I choose YOU! Pikachu!"
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From dl231 at eng.cam.ac.uk  Tue Aug  3 13:26:44 1999
From: dl231 at eng.cam.ac.uk (D. Lopez-De-Ipina)
Date: Tue, 03 Aug 1999 18:26:44 +0100
Subject: Question about serializing any IDL type objects in Python shelves
Message-ID: <37A72654.4BB565C1@eng.cam.ac.uk>

Hi,

I am trying to make a server store objects of IDL type
objectDetails(shown
bellow) in a Python shelve. Note the member propertyValue of structure
Property could represent any IDL type, even user defined IDL structures
that the TRIPServer doesn't know. I wonder if due to the dyanmic
properties of Python it would still be possible to serialize these
objects
or the CORBA 2.2 DynAny interface help would be required to achieve
that.

struct Property {
  string propertyName;
  any propertyValue;
};
typedef sequence<Property> PropertyList;

struct objectDetails {
  string label;
  PropertyList propertyList;
};

Thanks a lot,

Diego




From skip at mojam.com  Wed Aug 18 10:21:05 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 18 Aug 1999 09:21:05 -0500 (CDT)
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <m3672dwojr.fsf@pcrm.win.tue.nl>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<14265.47748.241386.291144@buffalo.fnal.gov>
	<wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
	<m3672dwojr.fsf@pcrm.win.tue.nl>
Message-ID: <14266.49377.51221.305982@dolphin.mojam.com>

    Stephan> Furthermore, Skip used he module whrandom, whereas I used
    Stephan> random.  Any ideas in which respect these modules differ?

Well, if you look at random.randrange, you'll find it exceedingly similar to 
whrandom.randrange:

    >>> import whrandom, random
    >>> random.randrange
    <method whrandom.randrange of whrandom instance at 814a370>
    >>> whrandom.randrange
    <method whrandom.randrange of whrandom instance at 814a370>

!

Skip



From gmcm at hypernet.com  Fri Aug  6 10:13:43 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 6 Aug 1999 09:13:43 -0500
Subject: Help with dictionary, please
In-Reply-To: <slrn7qlmht.35c.kas@titan.magnetic-ink.dk>
Message-ID: <1278177468-25600654@hypernet.com>

Klaus Alexander Seistrup wrote:

> The variable key should be a string, you're supplying a tupple.

Not at all. Tuples (or any immutable object) make fine dictionary 
keys.

The original problem is almost undoubtedly indentation (tabs & 
spaces, with editor interpretation out of sync with Python's).

- Gordon



From arcege at shore.net  Fri Aug 13 10:42:35 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 13 Aug 1999 14:42:35 GMT
Subject: Interesting thread on comp.lang.perl.misc
References: <37B42034.33F77AD5@strs.co.uk>
Message-ID: <v9Ws3.1713$EG4.332188@news.shore.net>

Ian Clarke <I.Clarke at strs.co.uk> wrote:
: There is an interesting thread discussing (in a more-or-less reasoned
: way) the relative merits of Perl and Python.  It's called "Why use Perl
: when we've got Python?".  Despite the provocative subject its not
: designed to start an advocacy war, and there are some very interesting
: comments being made about shortcomings in both languages.

: If you know both Perl and Python, I am sure your contribution would be
: welcomed.

Coming from a voice (in a choir) of experience, it does not go over well
when a lot of Python lovers post "we love Python because of..." in the
c.l.p.m newsgroup.  As you found, the Perl newsgroup is a different beast
from the Python newsgroup, but one thing is in common - we are each a
little defensive when ppl come to the newsgroup.

As you noticed from the postings (I had been reading the thread on
c.l.p.m), you had a number of people who were touting Perl and also
claiming they knew nothing about Python ("but that doesn't matter because
Perl is great").  You asked a perfectly valid question, just not in the
best way or to the best forum.

So what is the best forum?  Good question.  The best would be to find
papers by reputable persons who fairly judge the languages.  You got a
very good response from Tom C.  Read what he says; he's fair about both
languages.

  -Arcege




From frehe491 at student.liu.se  Thu Aug 12 14:30:30 1999
From: frehe491 at student.liu.se (Fredrik Henbjork)
Date: Thu, 12 Aug 1999 20:30:30 +0200
Subject: Copying Huge Files ? (Newbie!)
References: <37b30e52.43353218@news.usenetserver.com>
Message-ID: <99081220382305.14681@o112>

On Thu, 12 Aug 1999, you wrote:
> Hi,
> 
> 
> I want to copy some files across the disks of my server using python
> parse a script and then do what I want.
> 
> But while reading the manual I cannot find something like os.copyfile
> or os.movefile.
> 
> Are there some functions except open/read/write that can do the copy
> for me ???
> 
> 
> 
> Ciao
> 	Matthias Barmeier


Take a look at the things in shutil (especially shutil.copy and
shutil.copyfile) and os (especially os.rename and os.renames).

--
Fredrik Henbjork

Email: frehe491 at student.liu.se
WWW: http://o112.ryd.student.liu.se



From bruce_dodson at bigfoot.com  Thu Aug  5 22:37:06 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Thu, 5 Aug 1999 23:37:06 -0300
Subject: Any clues to source of this delay?
References: <al8vhaxtg1y.fsf@sirppi.helsinki.fi> <14247.22099.356207.503879@dolphin.mojam.com> <al8d7x4qkoc.fsf@sirppi.helsinki.fi>
Message-ID: <qXrq3.50753$jl.33906278@newscontent-01.sprint.ca>

This isn't unique to Python.  It has to do with the way TCP buffers data
within a time window by default.  The data is sent when some threshold size
is surpassed or when the time window expires (and I guess you saw that
sending zero bytes can force it to flush on some platforms).  For many
applications this buffering gives a performance improvement, but in your
case (the kind of application that wants to do lots round-trips per second
with small amounts of data in each request), it kills performance.

To optimize a socket for this kind of conversation, you can use socket
option TCP_NODELAY, but this is not a portable solution; I think it's a bit
different on Windows than on Linux, and it is not available on all
platforms.  I found no reference to TCP_NODELAY in the Python reference,
which is probably for the best given its non-portability.  If you are
targetting a platform that supports it, you can probably do something like:

  if (os.name == 'nt'):
    TCP_NODELAY = 1
    theSocket.setsockopt(IPPROTO_TCP, TCP_NODELAY, 1)

Depending on the nature of the conversation, one or both sides of your
connection may want to do that.  Protecting the nonstandard NODELAY option
with a platform test means the code will still work on other platforms,
although it will still have a "glass ceiling" on platforms that you haven't
planned for.

Bruce




From m.faassen at vet.uu.nl  Mon Aug  2 08:26:25 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 2 Aug 1999 12:26:25 GMT
Subject: Python microthreads
References: <FFropB.7tp@world.std.com>
Message-ID: <7o42ph$36i$1@newshost.accu.uu.nl>

Will Ware <wware-nospam at world.std.com> wrote:
[microthreads]

Okay, I am going to say not much at all, just that this sounds cool; I 
think this kind of thing fills a gap with Python. We have game programmers
etc asking for the same type of things as well, I think.

So keep up the good work!

Regards,

Martijn




From mwh21 at cam.ac.uk  Wed Aug 11 03:20:31 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Aug 1999 08:20:31 +0100
Subject: converting Netscrape mailbox to Unix mbox format
References: <14256.35316.328575.118034@lrz.de>
Message-ID: <m3wvv24wcg.fsf@atrus.jesus.cam.ac.uk>

Eugene Leitl <eugene.leitl at lrz.uni-muenchen.de> writes:

> Hi, I've inherited somebody's inbox as created by Her Royal Cruftiness
> Netscape. Does anybody have a Python script to salvage it back to
> hallowed standards of Unix?
> 
> TIA,
> 
> Eugene

Last I looked Netscape mailbox format *was* unix mbox format (on unix
anyway; can't speak for Netscape on windows) - have you tried pointing
pine at them?

Otherwise you could try the mailbox standard library module...

HTH
Michael



From tchrist at jhereg.perl.com  Thu Aug 19 13:21:10 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Thu, 19 Aug 1999 17:21:10 GMT
Subject: More random python observations from a perl programmer 
In-Reply-To: Ivan Van Laningham's message of Thu, 19 Aug 1999 11:19:55 -0600 .
             <37BC3CBB.7D28E3E5@callware.com> 
References: <37BC3CBB.7D28E3E5@callware.com>
Message-ID: <199908191721.LAA02333@jhereg.perl.com>

>> You're welcome.  I got so fed up with the damned flames coming
>> from people who don't actually even know what they're talking about,
>> that I decided to make sure I wasn't one of them. :-)
>> 

>If that's what you believe, why are posting stuff like this (from a
>different, but nearly simultaneous, posting)?

Stuff like what?

>> You seem to be looking for a fight.   Please say nothing
>> at all if that's all you want to do.

>In response to someone else's quite reasonable suggestion to read the
>documentation.


>You also say in the same message

>> Thanks for nothing: there are no manpages.  I read the Lutz book.
> 
>It's true that there are no manpages.  Linux comes with manpages, but
>every single manpage says (paraphrased): ``Don't trust this page; we
>don't.  See the info help system for the real scoop.''  

And they're full of crap, too.  Don't get me started.  Use BSD.

>Python doesn't have manpages, true.  It does have perfectly adequate
>documentation available at 
>	http://www.python.org/doc

>You can download that documentation and install it on your own computer
>if you wish to avoid the net.  Like all documentation, it leaves
>something to be desired (decent search facilities, for one, but manpages
>don't have those either); 

Yes, you can.  Well, I can.  I can show you how if you want.

I guess you'd better drop the manpage issue, because I don't
intend to stop considering its absence a nasty, glaring hole.

--tom



From dhellmann at zcentral.com  Mon Aug 23 10:00:37 1999
From: dhellmann at zcentral.com (Doug Hellmann)
Date: Mon, 23 Aug 1999 14:00:37 GMT
Subject: ANN: PmwContribD 1.11
Message-ID: <7prk5t$2ed$1@nnrp1.deja.com>

This library consists of several Pmw-based Tkinter widgets including a
ProgressMeter, basic TreeNavigator and a GUI application framework for
creating simple GUI apps quickly.

See http://members.home.net/doughellmann/PmwContribD/index.html for
more details.

(Those of you who may have looked at PmwContribD in the past please
notice that the web site has changed.)

Doug


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From skip at mojam.com  Thu Aug 26 11:52:15 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 10:52:15 -0500 (CDT)
Subject: binary data
In-Reply-To: <37C5169B.824BF1F3@xsoft.co.at>
References: <37C5169B.824BF1F3@xsoft.co.at>
Message-ID: <14277.25057.568284.823536@dolphin.mojam.com>

    Gerhard> How do I write binary data to a file?  I'm using open() and
    Gerhard> write and I always get the error that n is an integer.

I take it you're doing something like

    f = open("spam.bin", "w")
    n = 42
    f.write(n)

The file objects' write method only works with strings.  Take a look at the
struct module, which can pack the integer into a string:

    s = struct.pack("l", n)
    f.write(s)

If you don't really want the value written in binary, but in ascii, use the
% format operator instead:

    f.write("%d" % n)

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...




From jblake at stamp-coin.com  Tue Aug 17 17:41:01 1999
From: jblake at stamp-coin.com (Jonathon)
Date: Tue, 17 Aug 1999 21:41:01 +0000 (UTC)
Subject: Error Message: Can't assign to Operator
Message-ID: <Pine.LNX.4.04.9908172133450.24702-100000@natal.stamp-coin.com>

	I keep hitting this error message:
	SyntaxError:  can't assign to operator(line 3056)
	
	Line 3056 is
	meta-previous = string.join( the_ultimate_file_destination_is_this, string_record_id, suffix )
        
	the_ultimate_file_destination_is_this, string_record_id and  suffix 
	are all defined elsewhere.

	Questions:

	Why does that construct not work?
	How can I fix this construction?

	If you want the entire script, it is available at
	http://www.stamp-coin.com/python/script-error/index.html
	<< Actually, that points you to the file that can be
	downloaded. >>
	
	xan

	jonathon




From python-list at teleo.net  Sun Aug 29 23:46:22 1999
From: python-list at teleo.net (Patrick Phalen)
Date: Sun, 29 Aug 1999 20:46:22 -0700
Subject: Returning none
References: <14281.61423.566699.84062@dolphin.mojam.com>
Message-ID: <9908292048262F.02087@quadra.teleo.net>

On Sun, 29 Aug 1999, Skip Montanaro wrote:
> Martijn> Patrick Phalen <python-list at teleo.net> wrote:
>     >> On Sat, 28 Aug 1999, Alexander Williams wrote:
> 
>     >>> There are no procedures in Python.  There are only function calls (and
>     >>> Zuul).
> 
>     >> Zuul?
> 
>     >> Must be Dutch ...
> 
>     Martijn> It ain't Dutch. :)
> 
>     Martijn> Probably-some-PSU-codeword-ly yours,
> 
> Perhaps the lord of the underworld from Ghostbusters?

That's it! Thank you! That was bugging me.

"There is no Dana, only Zuul."



From news at dorb.com  Wed Aug 18 16:12:47 1999
From: news at dorb.com (Darrell)
Date: Wed, 18 Aug 1999 16:12:47 -0400
Subject: representation of unbound methods - strange behaviour
References: <37bbecf1.100354231@news.omnilink.de>
Message-ID: <7tEu3.37$ss2.2920@typhoon.nyroc.rr.com>

A variation that works.
>>> class A:
...     pass
...
>>> def f(self, x):
...     print x
...
>>> a=A()
>>> a.F=lambda x, self=a:f(self,x)
>>> a.F(55)
55
>>>

--
--Darrell
Stefan Franke <spamfranke at bigfoot.de> wrote in message
news:37bbecf1.100354231 at news.omnilink.de...
> Why does this happen:
>
> Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> class A: pass
> >>> def f(): pass
> >>> A.f = f
> >>> A.g = f
> >>> A.f
> <unbound method A.f>
> >>> A.g
> <unbound method A.f>
>                 ^^^
>
> This is independent of the assignment order of f and g:
>
> >>> class A: pass
> >>> def f(): pass
> >>> A.g = f        # swapped
> >>> A.f = f        # order
> >>> A.f
> <unbound method A.f>
> >>> A.g
> <unbound method A.f>
>
> Method caching?
> Stefan
>





From bruce_dodson at bigfoot.com  Fri Aug  6 14:08:04 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Fri, 6 Aug 1999 15:08:04 -0300
Subject: embedding
References: <7nstgb$st3$1@nnrp1.deja.com> <lxCq3.706$EG4.94343@news.shore.net>
Message-ID: <eAFq3.51249$jl.34326027@newscontent-01.sprint.ca>

>But, if you mean: "How can I create an object in the global namespace?"
>That is slightly different; for this you can use PyEval_GetGlobals().

>This returns a dictionary, on which you can then use
>PyDict_GetItemString() and PyDict_SetItemString().


BUT PyEval_GetGlobals will return null it it is not called from within
Python.  This will never be the case when extending Python, but when Python
is embedded, it may happen.  This is because PyRun_* define the global
scope, so if you're not somehow inside a PyRun_ thing (e.g. module called
from python called from embedder), there are no globals.

In an embedded-python situation, if you decide that all of your PyRun_*
calls are going to use __main__ as the global namespace,
PyModule_GetDict(PyImport_AddModule("__main__")) will do it.




From mcfletch at vrtelecom.com  Fri Aug 13 22:33:43 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Sat, 14 Aug 1999 02:33:43 GMT
Subject: Question about order in lists
In-Reply-To: <wx1t3.410$8P.14670@typhoon.nyroc.rr.com>
References: <wx1t3.410$8P.14670@typhoon.nyroc.rr.com>
Message-ID: <NDBBIAPHEKGOBIHCBLIIIEEDCAAA.mcfletch@vrtelecom.com>

Understandable behaviour:

l2 = l1[:] -> make a new list with size == l1 with pointers all pointing to
l1's data.

At the C level, this would mean approximately "copy memory range x to new
memory range y" (lists are contiguous arrays of pointers to objects).

Whereas:

for i in xrange(sz):l2.append(i)

Actually goes through x python loop iterations, where it binds the variable
i x times, finds attribute __getitem__ of your xrange x times, calls it x
times, finds the attribute append of l2 x times, and calls that function x
times.  (And every single variable lookup has the potential to fall through
to a higher-level namespace etceteras).  Python's dynamism has a _serious_
impact on performance and algo design ;) .

Oh, and the following is a fairly fast way to create your initial list of
integers (in this case, just a little faster):

range( 1000000 )

Cheers,
Mike

-----Original Message-----
From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
Behalf Of Darrell
Sent: August 13, 1999 7:06 PM
To: python-list at cwi.nl
Subject: Re: Question about order in lists


You win.
Map is very fast in this case.
Although something intresting happend when I used
l1[:] to create l2
##################
import operator, time

def mapIt(l1, l2):
    map( operator.add, l1,l2)


def loopIt(l1, l2):
    x=[]
    append=x.append
    for i in range(len(l1)):
        append(l1[i]+l2[i])


sz=1000000
l1=[]
for i in xrange(sz):l1.append(i)

if 0:
    # Time1: 5.75 sec
    # Time2: 27.70 sec
    l2=[]
    for i in xrange(sz):l2.append(i)
else:
    # Odd when using these huge numbers
    # This way of creating l2 ran faster ??
    # Time1: 3.25 sec
    # Time2: 28.50 sec
    l2=l1[:]

t1=time.time()
mapIt(l1,l2)
print " Time1: %.2f sec"%(time.time()-t1)
t1=time.time()
loopIt(l1,l2)
print " Time2: %.2f sec"%(time.time()-t1)

####################
# Time1: 3.25 sec
# Time2: 28.50 sec



--Darrell






From aahz at netcom.com  Sun Aug 15 15:31:50 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Aug 1999 19:31:50 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk>
Message-ID: <7p74j6$km0@dfw-ixnews5.ix.netcom.com>

In article <2GJsiHANGbt3Ew6i at jessikat.demon.co.uk>,
Robin Becker  <robin at jessikat.demon.co.uk> wrote:
>
>Lets face it; the perlers have a fairly hopeless position when it comes
>to defending the perl language (I once described perl as being like awk
>on steroids and lsd). 

Not really.  While there's lots to dislike about Perl, there's also some
things it does very nicely.  For example, it'd be really cool in Python
if I defined a function foo() and it could distinguish between each of
the following:

   x = foo()
   y = [ foo() ]
   z = { foo() }

Basic Perl is in some ways easier to pick up than Python for someone who
is a sysadmin but not a programmer.  Also, while it's easier to write
good Python programs, it's easier to debug Perl programs -- I'm not
counting IDLE because it requires a GUI.

Arguing over whether Python is "better" than Perl is like arguing over
whether European food is better than Asian.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From mlh at idt.ntnu.no  Thu Aug 26 09:58:58 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 26 Aug 1999 15:58:58 +0200
Subject: Tree module ?
References: <Pine.GSO.4.10.9908261547330.13215-100000@bonnie.sz-sb.de>
Message-ID: <y0jlnayve1p.fsf@vier.idi.ntnu.no>

Andreas Jung <ajung at sz-sb.de> writes:

> I'm looking for a general tree module for building non-binary unbalanced
> trees.=20
> 
> Any hints or comments ?=20

What kind of functionality do you require? If the building of trees is
all you need, there is hardly a need for a specific module...

There are *plenty* of ways of doing this... You could make some
fitting classes, or you could simply use lists... Each internal node
is a list of children, for instance...

> 
> Cheers,
> Andreas

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From chad at vision.arc.nasa.gov  Fri Aug 13 22:00:03 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Sat, 14 Aug 1999 02:00:03 +0000
Subject: Python Pocket Ref index available
References: <199908132316.RAA05367@shell.rmi.net>
Message-ID: <37B4CDA3.373885D7@vision.arc.nasa.gov>

Mark Lutz wrote:

> In response to reader requests, O'Reilly has just
> posted an index for the Python Pocket Reference, at:
>
>     http://www.ora.com/catalog/pythonpr/
>

And there was much rejoicing!  Thanks for the pointer, Mark.
(Although I can't connect to the server at the moment...)

Chad





From fredrik at pythonware.com  Fri Aug 13 09:28:30 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 13 Aug 1999 13:28:30 GMT
Subject: Anyone else find Tk akward?
References: <37B34DBB.E5A79760@sprint.ca>
Message-ID: <005001bee590$20581f50$f29b12c2@secret.pythonware.com>

Orlando Vazquez <ovazquez at sprint.ca> wrote:
> Is it just me, or is the way Tk organizes things on the window someone
> weird? I'm just starting to play around with it, and it has me puzzled.
> Wouldnt it be a lot easier to just use a coordinte system?
> For example put a button at (0,0) with length 10 and width 20 etc..

http://www.pythonware.com/library/tkinter/introduction/place.htm

> Are there any reasons why it _has_ to be this way?

yes.  try using place for a while, and you'll realize
why.

</F>




From faqtori at hotmail.com  Sun Aug 29 00:23:52 1999
From: faqtori at hotmail.com (Factory)
Date: Sun, 29 Aug 1999 14:23:52 +1000
Subject: Unsorting(randomizing) a sequence
References: <000401beebac$7a89b520$0c2d2399@tim> <m3d7wfm11h.fsf@pcrm.win.tue.nl>
Message-ID: <MPG.12317aab5a6714bc9896ea@news.melbpc.org.au>

In article <m3d7wfm11h.fsf at pcrm.win.tue.nl>, stephan at pcrm.win.tue.nl 
says...
> "Tim Peters" <tim_one at email.msn.com> writes:
> Actually, on Linux you could reseed by reading values from /dev/random.
> /dev/random will block if not enough "random" events are happening
> in the system to satisfy your application's hunger for random number.
> So this might actually make a little sense; a *real* random source
> (like /dev/random supposedly is) might be too slow to produce
> enough numbers.
> 
> ...although there is also /dev/urandom, which basically solves this
> problem for you (i.e. it uses the same mechanism as /dev/random to create
> random numbers, until the kernel is out of random numbers, and then
> it continues with pseudorandom numbers).

  Hmm this all might be made alot easier, since intel is apparently going 
to start including hw random generators on it's motherboards, hopefully 
every other comp. manufacturer will get the 'me too' feeling and start 
doing the same. 
  It'll prolly annoy alot of mathematician's though.. :)

  - Factory



From narf at golden.net  Thu Aug 26 18:14:51 1999
From: narf at golden.net (Matthew)
Date: 26 Aug 1999 18:14:51 -0400
Subject: no leading spaces in print
Message-ID: <87zozechpg.fsf@kloth.slux.dyndns.org>

I want to print, but I don't want the leading spaces
currently I am using 
print "whatever",
to have multiple print lines print on a single line

is there any way to fix this?



From kuncej at mail.conservation.state.mo.us  Tue Aug 24 11:05:53 1999
From: kuncej at mail.conservation.state.mo.us (Jeffrey Kunce)
Date: Tue, 24 Aug 1999 10:05:53 -0500
Subject: anyon know anythin about "squeeze"?
Message-ID: <s7c26f14.054@mail.conservation.state.mo.us>

>>a utility called squeeze 

>This program is in the source distribution "Tools/freeze" directory;

I see freeze, but not 'squeeze' in my 1.5.2 (I don't have CVS access).

The copy of squeeze that I use came from Fredrik Lundh at:
http://starship.skyport.net/crew/fredrik/ipa/squeeze.htm 

Chris Tismer made some enhancements with 'sqfreeze'
(I don't have a link handy for that - on starship somewhere I'm sure)

The latest development in this area seems to be
Gordon McMillan's python Installer.
http://starship.python.net/crew/gmcm/install.html


  --Jeff




From jeffp at crusoe.net  Sun Aug 15 17:28:53 1999
From: jeffp at crusoe.net (Jeff Pinyan)
Date: Sun, 15 Aug 1999 17:28:53 -0400
Subject: C's isprint() concept?
In-Reply-To: <Pine.GSO.4.10.9908151713380.3761-100000@crusoe.crusoe.net>
References: <slrn7rea88.9uf.jblaine@shell2.shore.net> <Pine.GSO.4.10.9908151713380.3761-100000@crusoe.crusoe.net>
Message-ID: <Pine.GSO.4.10.9908151727460.3761-100000@crusoe.crusoe.net>

>   $string =~ tr/\x00-\x1f\x80-\xff//d;

Perhaps, more readably:

  $string =~ tr/ -~//cd;

  the /c means complement (take the opposite of) the list
  the /d means delete those characters without a replacement
  the list is ' ' through '~', which is the class of printables

--
jeff pinyan    japhy at pobox.com  japhy+perl at pobox.com  japhy+crap at pobox.com
japhy's little hole in the (fire) wall:       http://www.pobox.com/~japhy/
japhy's perl supposit^Wrepository:       http://www.pobox.com/~japhy/perl/
The "CRAP" Project:                 http://www.pobox.com/~japhy/perl/crap/
CPAN ID: PINYAN           http://www.perl.com/CPAN/authors/id/P/PI/PINYAN/




From arcege at shore.net  Tue Aug 10 11:26:08 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 10 Aug 1999 15:26:08 GMT
Subject: Hidden CGI fields in HTML
References: <Pine.LNX.4.04.9908101441180.11952-100000@emerald.netskate.ru>
Message-ID: <kwXr3.1304$EG4.226668@news.shore.net>

Oleg Broytmann <phd at emerald.netskate.ru> wrote:
: On Tue, 10 Aug 1999, Ian Clarke wrote:
:> On a related issue, you are clearly experienced in using CGI with
:> Python.  I couldn't understand why cgi.FieldStorage returned a
:> dictionary which contained MiniFieldStorage objects (which must be
:> accessed using the .value field) rather than just the data itself.  This
:> would be much more convenient as I see it.  Any ideas?

:    There are, actually, some more issues on using FieldStorage, for
: example, you cannot ask for non-existing field:

: v = form["nofield"].value

:    will raise KeyError. And there is no way to emulate dict.get()...

Yes, you can
  if form.has_key("nofield"):
    if type(form["nofield"]) is type([]):
      nofield_value = map(lambda e: e.value, form["nofield"])
    else:
      nofield_value = form["nofield"].value
  else:
    nofield_value = None

But see below.

:    Oh, yes, I think I know why there is .value! What if your form has more
: than one value for some key? Like in the following GET:

: http://my.server/my/cgi/prog.py?city=Moscow&city=Paris&city=London

:    How do you get the value? How do you recognize thera are many values?
: Currently, you try to ask for form["city"], if the field exists you try to
: ask it for .value, but if the .value does not exists, you know it is a
: list:

:    for f in form["city"]:
:       print f.value

Actually, the reason is more of: what to do about file-uploads, where
you have the filename and the contents being given.  The FieldStorage
object will get a "filename" attribute, the opened file will be in
"file", and when you access "value" it will read the data from the
file.

To correct my code snippet above to handle files you have:
  # if nofield_value is of type:
  # None:  field not in form
  # list:  multiple values
  # tuple: tuple of the filename and file object
  # string: the string data from the form
  def get_field(form, fieldname):
    if form.has_key(fieldname):
      value = get_field_data(form[fieldname])
    else:
      value = None
    return value
  def get_field_data(field):
    if type(field) is type([]):
      # recursively handle the list (could have list of list of files)
      value = map(get_field_data, field)
    elif hasattr(field, "file"):
      value = (field.filename, field.file)
    else:
      value = field.value
    return value

I suggest reading the comments in Lib/cgi.py for more information.

  -Arcege




From anonymous at web.remarq.com  Tue Aug  3 15:05:35 1999
From: anonymous at web.remarq.com (anon)
Date: Tue, 03 Aug 1999 11:05:35 -0800
Subject: Python in web page
References: <37a70f28.96313821@midpoint>
Message-ID: <933707139.18310@www.remarq.com>

There are many Python resources on EarthWeb's new site,
www.opensource it.  Check out this page:
http://www.opensourceit.com/subjects/python.html  Hope this
helps!!



* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!



From mlh at idt.ntnu.no  Thu Aug 26 14:52:19 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 26 Aug 1999 20:52:19 +0200
Subject: Fuzzy string matching?
References: <199908261746.TAA25440@dionysus.fw.cuci.nl>
Message-ID: <y0jaerev0gs.fsf@vier.idi.ntnu.no>

"Hans Nowak" <hnowak at cuci.nl> writes:

> ...Anyone know of such a beast in Python?
>
[...] 
> Is such a thing done in Python already? If 
> not, would it be difficult to do? (I obviously don't how to do it 
> myself otherwise I would have done so already... :-/  )

I did some of this a while back, in a simple CGI-script trying to
merge two web pages... The point was - my local cinema had a web-page
showing which films were running, and when they were on etc. If you
clicked on the movie names, you invariably ended up on another page
which listed all the movie names - and if you clicked on one of
*them*, you would get a description of the film.

Idiotic design IMO.

So - I tried to make a simple script that extracted the URLs from the
second page and inserted them into the first, instead of the one
leading to the second... The problem was that the names weren't the
same...

There were several differences, among them: Different case,
abbreviations (& instead of "and"), different punctuation, different
cutoffs (length of strings), different use of whitespace, different
encoding of scandinavian letters (8-bit ASCII or HTML entities) etc.,
etc. And it was all unsystematic, as it seemed that the pages were
maintaned by different people...

So - I had a problem similar to yours... I didn't do anything
revolutionary (and the script is far from foolproof yet...), but...
What I did was remove all the things that were different, creating a
"standard form"... I removed all whitespace, made all letters
lowercase, changed all HTML entities into their ISO8859-1 equivalents,
cut the string after 20 characters, etc.

It worked to some degree. Of course, when you get misspellings etc.,
you have a problem... But you could use the (old) soundex module to
overcome some problems...

Other than that - fuzzy matching is a non-trivial problem. It is
possible to use dynamic programming to make a quadratic time distance
function that you can use to compare words or strings - to see how
similar they are. (If you like, I can post a Python variant of it.)

> 
> TIA,

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From nathan at islanddata.com  Thu Aug 12 20:36:56 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Thu, 12 Aug 1999 17:36:56 -0700 (PDT)
Subject: solaris 7--popen2 crash
Message-ID: <XFMail.990812173656.nathan@islanddata.com>

This is the first time I've had a test crash on me.  I've installed python
on linux, solaris, and HP-UX, but this is my first time on solaris 7. 
During make test, I got this:

test test_popen2 crashed -- exceptions.AssertionError : 

Any ideas?  Missing libraries?  Incompatible platform?  Any more helpful
information I can provide?  Worse case scenario: can I continue on using
python as long as I dont' use popen2? :)

----------------------------------
Nathan Clegg
 nathan at islanddata.com





From xdwvwz at hotmail.com  Tue Aug 24 16:31:33 1999
From: xdwvwz at hotmail.com (xdwvwz at hotmail.com)
Date: 24 Aug 1999 20:31:33 GMT
Subject: Unbelievable Opportunity Now  8554
Message-ID: <37c2f315.0@news.ocweb.com>

http://members.spree.com/business/sitecenter/
dgpwktezhxvbbbjudexkke




From jbauer at rubic.com  Wed Aug 18 20:12:51 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 19 Aug 1999 00:12:51 GMT
Subject: stat, chmod and RDONLY file (win32)?
References: <7pambk$3m8$1@ssauraaa-i-1.production.compuserve.com>
		<37B9D5F6.2AC782E1@compaq.com>
		<7pd13e$if1$1@ssauraab-i-1.production.compuserve.com> <37BB3AF1.53A9D306@compaq.com>
Message-ID: <37BB4C03.F6994282@rubic.com>

I've not been following this thread, but some time
ago I proposed some portable mechanism to simulate
Perl's file test operators:

http://www.python.org/cgi-bin/todo.py?req=show&file=faq04.012.htp

-Jeff

Greg Ewing wrote:
> "T. C. Mits" wrote:
> >
> > Funny I associated the four octal flags with Unix, not win98 or
> > NT.
> 
> The stat and chmod functions do their best to
> emulate the unix ones on other platforms.



From fredrik at pythonware.com  Fri Aug 20 08:17:03 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 12:17:03 GMT
Subject: Class Methods (as opposed to Instance Methods)
References: <7pjfhm$48p$1@feed.teaser.fr>
Message-ID: <001101beeb05$ea41d7a0$f29b12c2@secret.pythonware.com>

Olivier Deckmyn <olivier.deckmyn at mail.dotcom.fr> wrote:
> I play with Python while a few now (1 month).
> 
> My programer background is ObjectPascal, C++ and SmallTalk.
> 
> There is an "object-feature" I am not able to reproduce with Python : class
> method. As opposed to instance methods, theese should apply to classes.

python doesn't have class methods.  on the other
hand,  it makes very little sense to use them, when
you have to put all your code in modules anyway...

</F>




From tim_one at email.msn.com  Sun Aug 15 22:31:20 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 15 Aug 1999 22:31:20 -0400
Subject: Why use Python when we've got Perl?
In-Reply-To: <7p74j6$km0@dfw-ixnews5.ix.netcom.com>
Message-ID: <000301bee78f$6cd39de0$f22d2399@tim>

[Aahz Maruch]
> ...
> Arguing over whether Python is "better" than Perl is like arguing
> over whether European food is better than Asian.

Exactly so!  Everyone knows Asian food is better, but it's not polite to say
so in mixed company.

gastrowinkingly y'rs  - tim






From christian.egli at alcatel.ch  Thu Aug 12 13:43:45 1999
From: christian.egli at alcatel.ch (Christian Egli)
Date: 12 Aug 1999 19:43:45 +0200
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk>
Message-ID: <luemh851ym.fsf@stest.ch>

>>>>> "Phil" == Phil Hunt <philh at vision25.demon.co.uk> writes:

>>>>> "Paul" == Paul Prescod <paul at prescod.net> writes:

Paul> Glade is a GUI application that allows you to build a window
Paul> visually.
 
Phil> Okay, but I saw a reference somewhere to Glade having a
Phil> declarative file format that could be converted into code to
Phil> build the GUI. I was mostly talking about the declarative file
Phil> format.

Maybe you meant the storage format of Glade which is XML which in turn
can be used with Libglade to build the GUI dynamically.

Paul> Glade does that too, as long as your "favorite language" is
Paul> C. :)

Not only. See below

Paul> I don't understand why you would take a code generation
Paul> approach, though. My professional opinion is that code
Paul> generation is yucky. :) It takes no more effort to write a
Paul> "Parrot reader" for each language -- especially if Parrot uses
Paul> the language's built-in XML support (oops.  scratch
Paul> that). Seriously though, XML aside, why bother with code
Paul> generation when you can just do everything dynamically? Read the
Paul> parrot file and make the objects on the screen. The performance
Paul> shouldn't be much different and the extra dynamicity will make
Paul> your life easier and make the tool cooler to use. "I'll just
Paul> generate a modified parrot file and reload."

AFAIK Libglade does exactly that.

>From the Libglade home page:

Libglade is a library that performs a similar job to the C source
output routines in the GLADE user interface builder. Whereas GLADE's
output routines create C code that can then be compiled, libglade
builds the interface from an XML file (GLADE's save format) at
runtime. This way you can change the look of a program without needing
to recompile.

Currently it supports all the widgets in current releases, together
with support for keyboard accelerators and automatic signal
connection.

It is available from: ftp://ftp.daa.com.au/pub/james/gnome 


-- 
Christian Egli
Phone: +41 1 454-6728, Fax: +41 1 454-6612, http://www.stest.com
Wavetek Wandel & Goltermann, Friesenbergstr. 75, CH-8055 Zuerich



From fredrik at pythonware.com  Tue Aug 24 10:52:53 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 24 Aug 1999 14:52:53 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com> <7prsqk$9gr$1@nnrp1.deja.com>
Message-ID: <00f501beee40$58ebf950$f29b12c2@secret.pythonware.com>

<krodgers at tdyryan.com> wrote:
> > maybe we should add a note to the newsgroup FAQ?
> >
> > </F>
> 
> The comp.lang.fortran FAQ once had (maybe still has; I haven't looked at
> it in a while) an entry titled something like "Why are there useless
> debates?" that centered on a certain person at Purdue . . .

oh, I imagined something like "if you stumble upon some-
thing that looks like a flame, either ignore it altogether,
or work hard to make your answer pythonic rather than
pythogenic.  be extra careful if the poster is called
apollo, or has perl in the domain name".  or something.

</F>




From philh at vision25.demon.co.uk  Sun Aug 22 15:30:22 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 22 Aug 99 19:30:22 GMT
Subject: [Off-topic] Sysadmins (was Re: can some one help a newbie?)
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <935227166snz@vision25.demon.co.uk> <7pmv4j$gtt@dfw-ixnews14.ix.netcom.com> <935273107snz@vision25.demon.co.uk> <7pp2fk$7ul@dfw-ixnews9.ix.netcom.com>
Message-ID: <935350222snz@vision25.demon.co.uk>

In article <7pp2fk$7ul at dfw-ixnews9.ix.netcom.com>
           aahz at netcom.com "Aahz Maruch" writes:
> In article <935273107snz at vision25.demon.co.uk>,
> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> >In article <7pmv4j$gtt at dfw-ixnews14.ix.netcom.com>
> >           aahz at netcom.com "Aahz Maruch" writes:
> >> In article <935227166snz at vision25.demon.co.uk>,
> >> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> >>>In article <7pkke6$lut at dfw-ixnews6.ix.netcom.com>
> >>>           aahz at netcom.com "Aahz Maruch" writes:
> >>>>
> >>>> Overall, if your background is as a sysadmin using shell scripts, grep,
> >>>> awk, sed, and so on, you'll find Perl a bit easier.
> >>>
> >>>I've used all these, and find Python easier than Perl. Perl would
> >>>probably be easier than Python for 1-liners, if I could remember
> >>>the syntax.
> >> 
> >> Note that I'm specifically assuming a sysadmin with little actual
> >> programming experience.
> >
> >The idea that a company would employ someone, who can't really code,
> >as a sysadmin is IMO frightening. Unless they don't care whether
> >their computer system works, of course.
> 
> I guess I'm missing something, because the prospect doesn't "frighten"
> me.  There are a lot of activities for which a little bit of programming
> goes a long way, but I've seen a lot of people capable of doing that who
> really are not programmers in any sense of the word that you and I would
> use.  That's precisely a large part of why Perl has been so successful;
> it lends itself readily to the kind of "cookbook" programming that
> sysadmins do.

I think it depends on how complex your system is -- imagine a fairly 
complex setup, with hundered of little Perl scripts, all undocumented,
in different directories, depending on each other. 

-- 
Phil Hunt....philh at vision25.demon.co.uk




From bwarsaw at cnri.reston.va.us  Thu Aug 19 14:54:06 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 14:54:06 -0400 (EDT)
Subject: More random python observations from a perl programmer 
References: <37BC6986.A66B044E@starmedia.net>
	<199908191634.KAA01333@jhereg.perl.com>
Message-ID: <14268.21198.934117.241668@anthem.cnri.reston.va.us>

>>>>> "TC" == Tom Christiansen <tchrist at jhereg.perl.com> writes:

    TC> Thanks for nothing: there are no manpages.

You're right.  Is that a fatal lossage?  For some people, definitely
not for everybody, or even most Python programmers it seems (although
I do not follow the doc-sig, I haven't seen a tidal wave hue and cry
for manpages).

Is Python adequately documented?  Absolutely, because the latex source
is easily printed or viewed by a web browser (and you don't even need
to be on-line).  The current suite of documentation formats seems to
give poor overworked Fred the most bang for the buck.  Info format
used to be supported, but I don't think that was used by enough people
to justify the work involved in keeping it going.  Manpage format just
doesn't seem to be that popular either.  I know that sucks for you
Tom, but I'm sure Fred would gladly take a contribution that generated
manpage format from the latex source.  You might even be surprised at
the languages he'd accept such a contribution in! :)

It's really a simple matter of resources and the 80/20 rule.

-Barry



From mikael_aronsson at mail.bip.net  Sun Aug  8 14:58:59 1999
From: mikael_aronsson at mail.bip.net (Mikael Aronsson)
Date: Sun, 8 Aug 1999 20:58:59 +0200
Subject: Weird problem ?
Message-ID: <000601bee1d0$130870c0$adf9143e@pc58-mikael.fas.flygt.com>

Hi again !

You can forget about this problem, it was not related to Python, my
CRichEditCtrl messed up the string with some \r\n at the end of some lines,
and the Python parser did not like this at all.

Mikael






From bruce_dodson at bigfoot.com  Fri Aug 20 12:46:39 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Fri, 20 Aug 1999 13:46:39 -0300
Subject: twilight zone?  (re: random python observations...)
Message-ID: <GEfv3.69722$jl.44151689@newscontent-01.sprint.ca>

That was definitely a wierd debate.  I think we are being a bit too
defensive here.  Tom's observations are not off the mark.  Regarding
documentation, I find the tutorial to be "not bad" (though not enough to
forego buying Lutz's book), and the library  and API references to be pretty
good (though I wish it had a keyword search since I always find myself
groking the index).  I can't say the same for the PythonWin docs or the
pythoncom docs.  The COM documentation is too informal, and doesn't match
the style of the other documentation.  PythonWin's amount to three pages of
"preface" plus a sparse "quick reference appendix".  The whole middle of the
book is missing, which is why I'm waiting for Mark Hammond's rumored "Python
for Win32" book with bated breath.

Now before anyone shoots me, I know these extensions are not part of "core
Python", so it's not "Python's fault" and certainly not Fred's fault if
their documentation is not up to the same standards.  But maybe it is a
shortcoming of Python.  I don't know if anyone has installed perl, but its
documentation is just immaculate, in man form and also in HTML.  This
extends to the vast majority of the modules.  To a lesser degree, Java
packages also have good, consistent documentation.  This consistency in Java
and perl documentation is due, I think, to documentation facilities that are
included with the language distribution.

I guess no one saw the need for this in python because docstrings are built
in to the language (but you don't see the docstrings unless you're looking
at an interpreter window).  Or natural selection meant that people who cared
about consistent documentation were selected out of the Python community (or
Python was selected out of their vernacular, depending how you look at it).


The same goes for things like that (elt,) single element tuple.  That has
always bugged me.  I believe that the only reason " '%s' % aString " works
is because otherwise that trailing comma would have shown its ugly face in
more places than it does.  If the python syntax for creating a tuple didn't
look so much like the syntax for operator precedence, it would have been
better.
  "(a+5,)*2" looks too much like "(a+5)*2"
or (worse?):
  "(b+5.2)*3" vs. "(b+5,2)*3"
which is especially bad if you're French.


I've been thinking, maybe the Python community is just irreverant by nature,
owing to its namesake.  This can be a good thing, or it can be a bad thing.
In terms of diplomacy with other camps, I think it is turning out to be a
bad thing.  In terms of ignoring "old school" thinking and finding better
ways to do things, it's a good thing.  (Until we fall into the trap of
introducing our own dogma.  If we're going to be irreverant, we had better
be consistently irreverent, even toward ourselves!)

Bruce Dodson




From JamesL at Lugoj.Com  Sat Aug 28 15:00:13 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Sat, 28 Aug 1999 12:00:13 -0700
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net> <99082819072100.03655@o112> <7q98g5$7bi$5@newshost.accu.uu.nl>
Message-ID: <37C831BD.A5B17028@Lugoj.Com>

Martijn Faassen wrote:
> 
> Fredrik Henbjork <frehe491 at student.liu.se> wrote:
> > Let me set the facts straight: the Python Secret Underground movement does
> > not exist.  Every accusation that suggests that the PSU is, or was, involved
> > in, or associated with, the following things is completly false: the Kennedy
> > assassination, Richard Stallman,
> 
> Oh my god they KILLED Richard Stallman!!

Shucks, I knew that a long time ago. If you run emacs backwords it prints
out "Stallman is dead." The trick is running emacs backwords. Oddly enough,
it functions a lot like 'vi' when run backword.

Long live vi! (I had nothing to do with the virtual death of Stallman. And
he is really not dead; he just acts that way to fool people.)



From fredrik at pythonware.com  Sat Aug 14 09:30:03 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 14 Aug 1999 13:30:03 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cec7@cs.colorado.edu> <37b6185f.2442954@news.skynet.be>
Message-ID: <00cf01bee659$9ff9d220$f29b12c2@secret.pythonware.com>

Bart Lateur <bart.lateur at skynet.be> wrote:
> You're looking for the Holy Grail of orthogonality, again. Python is
> designed with orthogonality in mind, while Larry Wall allegedly once
> said:
> 
>     "Any computer scientist who praises orthogonality should be
>      sentenced to use an Etch-a-Sketch."

duh.  even the mainstream dictionaries can tell the
interested reader that "orthogonal" means more than
just "intersecting/lying at right angles".  for it's use
in computing, see FOLDOC.

</F>




From chinaski_py at my-deja.com  Sat Aug 21 06:01:41 1999
From: chinaski_py at my-deja.com (chinaski_py at my-deja.com)
Date: Sat, 21 Aug 1999 10:01:41 GMT
Subject: NEWBIE:Need Help with "Path"
Message-ID: <7plte3$c0t$1@nnrp1.deja.com>

This is my first post to this forum and I need help with the Python
path.I'm new to programming and I've been using PythonWin with no
trouble but when I try to open a .py file through Python it won't open.
I've read that I need to make sure that Python is in my "path".So I go
to(Windows98):
RUN
sysedit
Autoexec.bat
and this is literally the path:
path %path%;C:\Perl\bin;path=C:\PROGRA~1\Python\;%path%
       Can someone please tell me what I'm missing and/or doing wrong.
            Thanks,
              chinaski


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From skip at mojam.com  Thu Aug 19 14:48:30 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 19 Aug 1999 13:48:30 -0500 (CDT)
Subject: Python to Perl Conversions
In-Reply-To: <7phiek$sn4$1@nntp6.u.washington.edu>
References: <37bc1303@cs.colorado.edu>
	<slrn7ro9ns.h6v.jblaine@shell2.shore.net>
	<37bc2ee5@cs.colorado.edu>
	<slrn7rog4u.hri.jblaine@shell2.shore.net>
	<7phiek$sn4$1@nntp6.u.washington.edu>
Message-ID: <14268.20820.108485.396423@dolphin.mojam.com>

    Donn> This core stability is itself a virtue of Python, but I wonder if
    Donn> it has inhibited the development of the documentation.

It's probably reduced the number of gratuitous Python books and
updates... ;-)

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From wtanksle at dolphin.openprojects.net  Wed Aug 25 12:49:43 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Wed, 25 Aug 1999 16:49:43 GMT
Subject: xanadu and python
References: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com>
Message-ID: <slrn7s87l6.l8j.wtanksle@dolphin.openprojects.net>

On Wed, 25 Aug 1999 08:03:19 GMT, Fredrik Lundh wrote:
>I doubt anyone have missed that Ted Nelson's Xanadu,
>the original hypertext system, has finally been released
>as open source:

>    http://www.udanax.com/

>what Python has to do with it?  well, go to the site and
>find out for yourself ;-)

A better URL to go to first is http://www.scripting.com/, which has the
working mirror as well as links to the Python code on the web.  And it's
some clean code, too -- if Xanadu is as cool as it sounds, this might be a
killer app for Python.

And best of all, it's a truly unencumbered license!

></F>

-- 
-William "Billy" Tanksley



From fdrake at acm.org  Thu Aug 26 16:24:18 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 26 Aug 1999 20:24:18 GMT
Subject: Python article on Builder.com
In-Reply-To: <MPG.122f52adbcf97bbd989698@news.nettally.com>
References: <slrn7rpqbj.atl.kjz@kzaragoza.ne.mediaone.net>
	<MPG.122f52adbcf97bbd989698@news.nettally.com>
Message-ID: <14277.41586.877892.694820@weyr.cnri.reston.va.us>

Tom Funk writes:
 > Dan paints a nice picture of Python, but the discussion needs some input 
 > from some experienced Python folks (Andrew? Tim? Barry? Fred? Anyone?).

  Oh! Oh!  Is that me??  Did I make the list?!
  Nah, must've meant Fredrik.  ;-)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From kranio at nospam.nospam.it  Tue Aug 31 08:48:02 1999
From: kranio at nospam.nospam.it (Kranio)
Date: Tue, 31 Aug 1999 12:48:02 GMT
Subject: calling python interpreter
Message-ID: <37cbccd8.5014784@news.tiscalinet.it>

Under Red Hat 5.x I use to call the file pippo.py as

$ python pippo.py

I would like to call it as

$ ./pippo.py 

by writing the first line

#!/usr/bin/python

but it gave me "access denied" message. The interperter is in the
right dir and I am the owner of the file; tha same happens being root.
What is the problem?

Marco Buzzo



From bowman at montana.com  Fri Aug 20 00:08:05 1999
From: bowman at montana.com (bowman)
Date: Thu, 19 Aug 1999 22:08:05 -0600
Subject: No man pages? (was Re: More random python observations from a perl programmer)
References: <199908192043.OAA07181@jhereg.perl.com>
Message-ID: <37BCD4A5.482CC246@montana.com>

Tom Christiansen wrote:
> 
> That there appears to be running a 1.5.1 version on a Redhat
> Linux system.  I haven't tried a from-scratch build.  I'm
> happy to blame RH.  Their doc situation is nasty.

Would that be the same RH5.2 that caused me a day's frustration because
their
Perl is so broken it can't handle PerlTk? You want to be careful
building Python 1.5.2, also, as a little misstep will break RH's
'glint', 'linuxconfig' and other Python based oddities.

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From loewis at informatik.hu-berlin.de  Fri Aug 13 13:05:26 1999
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 13 Aug 1999 19:05:26 +0200
Subject: Does Python run under Windows 2000? (build 2031)
References: <37B40D58.DD6AADC0@RobotVision.com>
Message-ID: <p6qpv0robl5.fsf@informatik.hu-berlin.de>

"James E. Green" <jgreen at RobotVision.com> writes:

> I'm running Windows 2000 RC1 (build 2031, the very latest).  Will Python
> run under this OS?

It works for me. You'd hope M$ would provide that much compatibility.

Regards,
Martin



From arcege at shore.net  Thu Aug 12 12:28:20 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 12 Aug 1999 16:28:20 GMT
Subject: Why is CGI module so complicated?
References: <37AB04A8.C3403427@NOSPAM.strs.co.uk> <cWYr3.1329$EG4.226668@news.shore.net> <37B15BF8.B7204705@strs.co.uk> <mjhs3.1462$EG4.260559@news.shore.net> <y0jiu6lf5q7.fsf@drue.idi.ntnu.no>
Message-ID: <ECCs3.1608$EG4.299767@news.shore.net>

Magnus L. Hetland <mlh at idt.ntnu.no> wrote:
: "Michael P. Reilly" <arcege at shore.net> writes:

:> Ian Clarke <I.Clarke at strs.co.uk> wrote:
: [...]

:> 
:> What would be _really_ good (IMO) is to have a "toDict" method that
:> does this.

: Why not just a factory function in the cgi module?

Because (IMO):

a.  It would probably be better to have only one interface; especially
when that interface(s) is to instantiate the cgi input only once,
during a program.  Remember that POST methods don't allow you to
recreate the form data a second time (it assumes standard input, which
is unseekable).  This is a lamer reason than below, but clearer-cut
interfaces are part of what makes Python better than some other
languages (when it's set up like that :).

b.  The method is more usable than a factory function called instead of
FieldStorage.  You mention being able to use this with TemplateDocument
classes (in another article);  what if you want to get access to ALL
the data, but also want to give a dictionary representation to some
other interface as well?  The factory function returns the dict and
throws away the rest of the data, whereas creating the class preserves
all the data you need, keeping it available in your app, but also can
give the dictionary you need at other times.

BTW, I also like the idea of adding the __str__ method to
FieldStorage.

:   Magnus              Making no sound / Yet smouldering with passion
:   Lie          The firefly is still sadder / Than the moaning insect
:   Hetland                                       : Minamoto Shigeyuki

  -Arcege




From bkc at Murkworks.com  Mon Aug  2 15:06:13 1999
From: bkc at Murkworks.com (Brad Clements)
Date: Mon, 2 Aug 1999 15:06:13 -0400
Subject: C++ wrappers around the C API?
References: <rqbmgaa0f8fcq6@corp.supernews.com>
Message-ID: <7o4q8a$78m$1@news.clarkson.edu>

The LLNL distribution already has a start in this direction.

See http://xfiles.llnl.gov/CXX_Objects/cxx.htm




--
Brad Clements,
bkc at murkworks.com
Rich Salz <salzr at certco.com> wrote in message
news:rqbmgaa0f8fcq6 at corp.supernews.com...
> Has anyone written a set of C++ wrappers around Python's C API?
> Is there any interest in this?
>
>
>





From andrew at starmedia.net  Fri Aug 20 16:20:28 1999
From: andrew at starmedia.net (Andrew Csillag)
Date: Fri, 20 Aug 1999 20:20:28 GMT
Subject: Setting environment variables.
References: <37BD3429.7364B4F7@iinet.net.au>
Message-ID: <37BDB88C.212EC063@starmedia.net>

Oliver White wrote:
> 
> I'm writing my first python program...
> 
> The program is intended to give a menu of choices, from which the
> CVSROOT environment variable is set (I'm sick of swapping around
> manually for sucking down code from different places). I'm wondering how
> I would do that.
> 
> I'm running GNU/Linux, is there an issue with processes... eg, after
> running my python program as a child process, will the changes to the
> environment variable affect it's parent (the *term)? (suspecting no) Is
> there annother way to set the parent's?

You would be correct, changing environment variables in a child process
won't affect the parents environment variables.  A way to do it is have
the child process print the new value of the environment variable to
stdout and have the parent read this value (probably using os.popen or
similar) and set its environment accordingly.
-- 
"Programmers are just machines that convert coffee to executable code"
   -Unknown

"Drew Csillag" <drew_csillag at geocities.com>



From bboard at parc.xerox.com  Fri Aug 13 16:47:12 1999
From: bboard at parc.xerox.com (Mail Archiving Daemon)
Date: Fri, 13 Aug 1999 13:47:12 PDT
Subject: open file errors in Python?
Message-ID: <99Aug13.134712pdt."3510"@watson.parc.xerox.com>

I notice that when I set up a simple loop to open files in Python,
it bombs out after opening 251 files.  My fd budget is more like 1024.
Why and where does Python impose this lower limit?  Also, why is the
errno signalled 0?  Instead of EMFILE?

Bill



From -  Sat Aug  7 15:22:26 1999
From: - (-)
Date: 7 Aug 1999 19:22:26 GMT
Subject: Home Based Business
Message-ID: <7oi11i$77m2342@news.emirates.net.ae>

Earn unlimited income working at home !
Start your own e-mail business today.
Starting your own e-mail business will enable you
to take advantage of the following:
1. Top part-time owners have earned $5000+ in one month.
2. Top full-time owners have earned $10000+ in one month.
3. All customers pay you in cash.
4. You will sell a product which costs nothing to produce.
5. Your only overhead is your time.
6. Your startup costs are less than $25
7. You have more than 40 million potential customers.

For additional information e-mail me at   mria098 at yahoo.com



From fredrik at pythonware.com  Mon Aug 30 03:53:43 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 30 Aug 1999 07:53:43 GMT
Subject: tkinter/python question
References: <7q80gl$t10$1@hermes.louisville.edu>
Message-ID: <014e01bef2bd$3f8edaa0$f29b12c2@secret.pythonware.com>

Jeff -=Machtyn=- Gosnell <jmgosn01 at athena.louisville.edu> wrote:
> I have a question concerning tkinter and python.  Which type of user is
> tkinter built for?

the kind of programmer who's writing Python code,
and wants a simple and portable GUI toolkit.

...

Tcl doesn't really have anything to do with it, except
that (1) the Tcl interpreter is linked with Tkinter, and
(2) there are no complete Tkinter documentation out
there just yet, so you sometimes have to resort to the
Tk manual pages, which happens to use Tcl syntax...

the former is a technical detail; the latter is being
worked on, on several fronts.  my freely available
tkinter introduction includes references pages for
most widgets:

http://www.pythonware.com/library.htm

and there are (at least) two Tkinter books in the
works; one by John Grayson (for Manning) and one
by yours truly (for O'Reilly).

</F>




From graham at sloth.math.uga.edu  Sun Aug 22 21:45:50 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 23 Aug 1999 01:45:50 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <37BEC755.645A350D@fnmail.com>
Message-ID: <7pq94e$3gd$1@cronkite.cc.uga.edu>

Per Kistler (kistler at fnmail.com) wrote:
: ...but Tom has been attacked over and over again. I can really
: understand that he may loose temper after beeing treeted like
: a idiot again and again. And Tom is really not any dummy. 
: I suggest to treat the "bigger" guys in the computer community
: with more respect

Why should I treat a "bigger guy" with more respect? If the "bigger
guy" presents a crap argument he gets all the respect he deserves,
namely none.

graham

-- 
              Where do you think you're going
              Don't you know it's dark outside
              Where do you think you're going
              Don't you care about my pride



From tchrist at jhereg.perl.com  Thu Aug 19 13:13:52 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Thu, 19 Aug 1999 11:13:52 -0600
Subject: More random python observations from a perl programmer 
In-Reply-To: Jeremy Hylton's message of Thu, 19 Aug 1999 13:14:09 -0400 (EDT) .
             <14268.15201.81468.481452@bitdiddle.cnri.reston.va.us> 
Message-ID: <199908191713.LAA02152@jhereg.perl.com>

>Indeed, I enjoyed reading both your posts this morning.  While I don't 
>agree with all of your criticisms, I think you make many good points.
>I did notice that some of your criticisms were out-of-date, perhaps
>reflecting Python previous to 1.5.2.  

I only have the Lutz book.  I don't have manpages. :-(

--tom



From urner at alumni.princeton.edu  Tue Aug 24 11:03:34 1999
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Tue, 24 Aug 1999 15:03:34 GMT
Subject: Pro Python remarks to math ed folks (pointers)
References: <37c2a51e.18050990@news.teleport.com> <Ibjw3.680$v6.61454@typhoon.nyroc.rr.com>
Message-ID: <37c2b213.49798667@news.teleport.com>

"Darrell" <news at dorb.com> wrote:

>Installed your stuff and changed a few things to get it to work.
>Had to add this method too Vector.
>   def display(self):
>       print self.data

That's cool.

I've just been going:

from vector import Vector
v = Vector(quad=(1,0,0,0))
print v.data
(1.0, 0.0, 0.0, 0.0)

You have xyz=3-tuple, quad=4-tuple and polar=3-tuple as your
three input options.  When you rotate, you go:

v.rotate((0,0,30)) # 30 degrees around Z axis

or

v.rotate((10,0,0,0)) # 10 degrees around the A axis

This 'quad' stuff (with 4-tuples) is the most unusual and
exotic and would stump any classroom math teacher for a few
minutes because it's unfamiliar and not taught as a normal
part of any curriculum.  But that doesn't keep "quadrays"
from being useful and fun.  The polyhedra I'm interested in
are very easily constructed using these non-traditional 
basis vectors, to the 4 corners of a regular tetrahedron
from an origin at (0,0,0,0) -- tetrahedron's center.

Note that ps module has lots of test scripts and does the
importing it needs, i.e.

import ps
ps.testprog2()

will write out some files (either POVray, VRML or both).

>Also changed some of the hardcoded paths.

Yes, ps has some of those.  Good point.

>Then installed a VRML viewer.
>I'm digging around for a POV viewer still.

www.povray.org is where to get Povray.  It's really a rendering
process, computational in nature, not just viewing, i.e. all
the Python supplies is ASCII text, with Povray has to munch
and crunch (same with VRML of course).

>Wow, very cool.
>

Thanks for the positive feedback.

>I have a teenager who hates math, maybe this would be useful.

I sure think there's a lot of potential here, especially when
you feel fluent enough to start doing art projects with 
polyhedra.  My website has lots of Povray graphics I did 
using this Python or earlier Java or Xbase incarnations of
this application.

Examples: 
http://www.inetarena.com/~pdx4d/ocn/oop6.html
http://www.inetarena.com/~pdx4d/ocn/oop5.html

Animated GIFs: http://www.inetarena.com/~pdx4d/ocn/xtals101.html

Kirby




From pehr at alum.mit.edu  Fri Aug  6 00:53:17 1999
From: pehr at alum.mit.edu (pehr anderson)
Date: Fri, 06 Aug 1999 00:53:17 -0400
Subject: Get 'ifconfig' information via Python
References: <87yafs8ruu.fsf@linuxcare.com> <37A81400.7474FD89@lemburg.com>
Message-ID: <37AA6A3D.7A33C95F@alum.mit.edu>

Dear David,
I really like the idea of writing an extension module.
I was writing Java aplications under windows and
ran into a complete brick wall because there was
no "inherent" way to get this data with the defined APIs.
There was just no expectation for multi-homed systems
as far as I could tell.

Please add me to your list of interested parties, if you
are compiling such a thing.
I've never written an extension module for python
but have done several python scripts which grabbed the relevant
numbers out of os.popen("ifconfig").read()
I'm really interested in seing this solved "the right way".
Perhaps a request should be made on cosource.com?

This  is a problem that needs a real cross-platform solution.
      -pehr


"M.-A. Lemburg" wrote:

> David N. Welton wrote:
> >
> > So, I think I'm getting close...
> >
> > import fcntl
> > import IN
> > import struct
> >
> > s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> >
> > res = fcntl.ioctl(s, IN.SIOCSIFHWADDR, ??????)
> >
> > I can't seem to find what goes in ?????.
> >
> > Is this the right way to be doing things?  I suppose I can always go
> > pull apart the sources for ifconfig itself, and make a .so that does
> > what I need, but I get the feeling that I can do what I need from
> > within Python.
> >
> > Yeah, I could popen ifconfig, too, but that's ugly.
> >
> > Downloading python sources now to see what I can see in the fcntl
> > module...
>
> The .ioctl() function can take a string or an integer as
> third argument. To find out what to pass for a given option,
> look at man ioctl and man ioctl_list. The exact meaning is
> not mentioned there unfortunately, though.
>
> On Linux, just dig into the net/core/dev.c file to find out
> what happens for the above option:
>
> The argument is being interpreted as struct ifreq and then copied
> to ifr...
>
>                 case SIOCGIFHWADDR:
>                         memcpy(ifr.ifr_hwaddr.sa_data,dev->dev_addr, MAX_ADDR_LE
>                         ifr.ifr_hwaddr.sa_family=dev->type;
>                         goto rarok;
>
>                 case SIOCSIFHWADDR:
>                         if(dev->set_mac_address==NULL)
>                                 return -EOPNOTSUPP;
>                         if(securelevel > 0)
>                                 return -EPERM;
>                         if(ifr.ifr_hwaddr.sa_family!=dev->type)
>                                 return -EINVAL;
>                         ret=dev->set_mac_address(dev,&ifr.ifr_hwaddr);
>                         break;
>
> The basic ifreq struct is defined in include/linux/if.h:
>
> /*
>  * Interface request structure used for socket
>  * ioctl's.  All interface ioctl's must have parameter
>  * definitions which begin with ifr_name.  The
>  * remainder may be interface specific.
>  */
>
> struct ifreq
> {
> #define IFHWADDRLEN     6
> #define IFNAMSIZ        16
>         union
>         {
>                 char    ifrn_name[IFNAMSIZ];            /* if name, e.g. "en0" *
>         } ifr_ifrn;
>
>         union {
>                 struct  sockaddr ifru_addr;
>                 struct  sockaddr ifru_dstaddr;
>                 struct  sockaddr ifru_broadaddr;
>                 struct  sockaddr ifru_netmask;
>                 struct  sockaddr ifru_hwaddr;
>                 short   ifru_flags;
>                 int     ifru_metric;
>                 int     ifru_mtu;
>                 struct  ifmap ifru_map;
>                 char    ifru_slave[IFNAMSIZ];   /* Just fits the size */
>                 caddr_t ifru_data;
>         } ifr_ifru;
> };
>
> #define ifr_name        ifr_ifrn.ifrn_name      /* interface name       */
> #define ifr_hwaddr      ifr_ifru.ifru_hwaddr    /* MAC address          */
> #define ifr_addr        ifr_ifru.ifru_addr      /* address              */
> #define ifr_dstaddr     ifr_ifru.ifru_dstaddr   /* other end of p-p lnk */
> #define ifr_broadaddr   ifr_ifru.ifru_broadaddr /* broadcast address    */
> #define ifr_netmask     ifr_ifru.ifru_netmask   /* interface net mask   */
> #define ifr_flags       ifr_ifru.ifru_flags     /* flags                */
> #define ifr_metric      ifr_ifru.ifru_metric    /* metric               */
> #define ifr_mtu         ifr_ifru.ifru_mtu       /* mtu                  */
> #define ifr_map         ifr_ifru.ifru_map       /* device map           */
> #define ifr_slave       ifr_ifru.ifru_slave     /* slave device         */
> #define ifr_data        ifr_ifru.ifru_data      /* for use by interface */
>
> Building these structs can be done using the Python struct
> module... playing around with this can probably crash your system
> though.
>
> In the end, I think you're better off hacking together a
> new extension module.
>
> Hope that helps.
>
> --
> Marc-Andre Lemburg
> ______________________________________________________________________
> Y2000:                                                   149 days left
> Business:                                      http://www.lemburg.com/
> Python Pages:                           http://www.lemburg.com/python/




From 101665.216 at compuserve.com  Tue Aug 24 11:32:03 1999
From: 101665.216 at compuserve.com (Robert Land)
Date: Tue, 24 Aug 1999 15:32:03 GMT
Subject: UML
References: <3F8BA119B48AD11197470008C7A493CB98B502@HQOBGEX1>
Message-ID: <37c2afa8.67206110@news.compuserve.com>

Thanks for your reply Bill.
I'm afraid I still cannot imagine what the UML looks like - I cannot
take hold of it. Remarkably almost everyone wrote simular
explainations - so there isn't much more to talk about without losing
the global view.
I'll try it the other way around. Is the UML something like a OO
Database or something like XML? Or anything in between?

I tried variuos links on the Internet, some were dead - but the most
led to a book advert. Two books seem to suite for my poor knowledge in
this area:


1.) Object Primer(Scott W. Ambler SIGS Books/
     Cambridge University Press, 1995)
2.) UML in a Nutshell (OReilly)

Which - or would you recommend both?

I started documenting my work on AI in a simple word processor, then
thinking of creating a database (to get hold of the code and my
thoughts) and recently I heard something about CASE tools and UML.
I'm really in great need of a tool which helps me to keep overview and
organize my work.

Some aditional questions:

On Tue, 17 Aug 1999 13:17:30 -0400, ScherBi at BAM.com wrote:

>I use UML, via ObjectDomain (www.objectdomain.com).
>I am involved in developing ....
>.. In other words, create classes and class diagrams in
>UML, design the class interfaces, and play around with the design at this
>level.  When you think you have it right, generate the stub code and fill in
>the methods.  You can fill in the methods without getting caught up in the
>interface issues, which makes things much easier.

Feels like UML is a sort of IDE??


>..where you understand the hierarchy and the operations, then
>reverse engineer it into objectdomain.  You can then automatically generate
>a class diagram.  

Something like a class browser used in C/C++ IDE's like Watcom?


>And no, I don't work for ObjectDomain.  It's the only UML tool I know of the
>generates and reverse engineers Python.  OD is written in Java.  I don't

Whats so wrong about Java? When you use different OS's - Java should a
logical way to go.


Thanks for your patience


Robert   

*(kopy of email and news followup are identical)




From xdsi at xddainese.unive.it  Tue Aug 17 18:08:12 1999
From: xdsi at xddainese.unive.it (Diego Dainese)
Date: 18 Aug 1999 00:08:12 +0200
Subject: How to write to a named pipe?
References: <7p87nv$fbl$1@nnrp1.deja.com>
Message-ID: <87ogg6giwj.fsf@blanka2.blankanet.it>

singlets at my-deja.com writes:
[snip]
> %: python
> Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC
> egcs-2.91.66 19990314/Li on linux-i386
> Copyright 1991-1995 Stichting Mathematisch
> Centrum, Amsterdam
>>>> import os
>>>> os.mkfifo('/tmp/myfifo')
>>>> f=open('/tmp/myfifo','w')    # now we're stuck
[snip]

It is normal that Python hangs when you try to open the fifo for
writing, because the underlying C sys-call hangs waiting for some
process to open the pipe for reading.

The problem is that when Python hangs, it cannot receive signals; I
had the very same problem with the threading API, and the only
solution I found was to do polling.

Does anyone known a better solution?

> Sorry if I'm being an idiot.
> 

I'm sorry too ;)

Bye,

-- 
Diego   | To reply remove the `x' and swap the | Sorry for my
Dainese | words around the `@' in the address. | bad English!



From fredrik at pythonware.com  Fri Aug 20 10:02:17 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 14:02:17 GMT
Subject: Reading from serial port and dumping to console
References: <37BD20F8.9A984D7E@pc.jaring.my>
Message-ID: <00ef01beeb14$9d6cd060$f29b12c2@secret.pythonware.com>

Thomas Lim <englim at pc.jaring.my> wrote:
> I'm trying write a program to read from the serial port (barcode reader
> attached) and dump to the console, so that my application program can
> receive the data.
> OS: Red Hat Linux 6
> Python: 1.5.2
> 
> Program listing:
> p = open('/dev/ttyS0', 'r')
> t = open('/dev/tty0', 'w')
> while 1:
>     x = p.readline()
>     print x   # yes, I could read from /dev/ttyS0
>     t.write(x)  # but this has no effect on console
> 
> Hope that someone could help.

how about:

p = open('/dev/ttyS0', 'r', 0)
t = open('/dev/tty0', 'w', 0)

(0 means unbuffered I/O)

hope this helps!

</F>




From c.curtis at worldnet.att.net  Sun Aug  1 13:31:05 1999
From: c.curtis at worldnet.att.net (Craig Curtis)
Date: Sun, 1 Aug 1999 13:31:05 -0400
Subject: spawnv
Message-ID: <19990801171952.JFVG5731@default>

I am trying to spawn c:\windows\notepad.exe from a Python program.  I want
the python script to wait for Notepad to exit.  
How can I do this?  

I have tried all combinations I can think of using execv, spawn, system and
cannot get it to work.

Could someone please post the exact command sequence to type in.

Thanks in advance,
Craig Curtis



From wtanksle at dolphin.openprojects.net  Tue Aug 24 19:13:40 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 24 Aug 1999 23:13:40 GMT
Subject: UML
References: <3F8BA119B48AD11197470008C7A493CB98B502@HQOBGEX1> <37c2afa8.67206110@news.compuserve.com>
Message-ID: <slrn7s69p4.jb7.wtanksle@dolphin.openprojects.net>

On Tue, 24 Aug 1999 15:32:03 GMT, Robert Land wrote:
>Thanks for your reply Bill.
>I'm afraid I still cannot imagine what the UML looks like - I cannot
>take hold of it. Remarkably almost everyone wrote simular
>explainations - so there isn't much more to talk about without losing
>the global view.

UML is a bunch of conventions for drawing pictures which illustrate your
design.

>I'll try it the other way around. Is the UML something like a OO
>Database or something like XML? Or anything in between?

Nothing like any of that.  It's like the stencils they sell in drafting
shops.

>I tried variuos links on the Internet, some were dead - but the most
>led to a book advert. Two books seem to suite for my poor knowledge in
>this area:

>1.) Object Primer(Scott W. Ambler SIGS Books/
>     Cambridge University Press, 1995)
>2.) UML in a Nutshell (OReilly)

>Which - or would you recommend both?

I've never seen the first, but I tried to learn from the second and HATED
it.  The author seems to delight in using bullet lists as part of every
paragraph, and new terms complete with definition (sometimes) are brought
in on every page.  I was _really_ disappointed.

If you already know UML, there's a possibility that it would work for you.
If you don't, there's very little chance.

>I started documenting my work on AI in a simple word processor, then
>thinking of creating a database (to get hold of the code and my
>thoughts) and recently I heard something about CASE tools and UML.
>I'm really in great need of a tool which helps me to keep overview and
>organize my work.

UML simply helps you lay out your design on paper.

I don't have any other recommendations.

>Some aditional questions:

>On Tue, 17 Aug 1999 13:17:30 -0400, ScherBi at BAM.com wrote:

>>I use UML, via ObjectDomain (www.objectdomain.com).
>>I am involved in developing ....
>>.. In other words, create classes and class diagrams in
>>UML, design the class interfaces, and play around with the design at this
>>level.  When you think you have it right, generate the stub code and fill in
>>the methods.  You can fill in the methods without getting caught up in the
>>interface issues, which makes things much easier.

>Feels like UML is a sort of IDE??

No, but ObjectDomain (which I hadn't seen before) uses UML to help
structure your program.  I'll have to check it out.

>>And no, I don't work for ObjectDomain.  It's the only UML tool I know of the
>>generates and reverse engineers Python.  OD is written in Java.  I don't

>Whats so wrong about Java? When you use different OS's - Java should a
>logical way to go.

Python runs on more OSes.  Python even runs on Java.

Java isn't that nice of a language -- all you can say is that it's better
than C++.

>Thanks for your patience

>Robert   

-- 
-William "Billy" Tanksley



From philh at vision25.demon.co.uk  Thu Aug 26 16:33:04 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Thu, 26 Aug 99 20:33:04 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com> <37C56146.7709B260@megsinet.net>
Message-ID: <935699584snz@vision25.demon.co.uk>

In article <37C56146.7709B260 at megsinet.net>
           davidopp at megsinet.net "David Oppenheimer" writes:
> Perhaps tchrist should start a group called "comp.lang.flames" ?

Good idea, I think ``Flames'' wouyld be an excellent name for a new 
language.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From skaller at maxtal.com.au  Wed Aug  4 02:30:52 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Wed, 04 Aug 1999 06:30:52 GMT
Subject: 1.5.2 for: else:
References: <1278928389-16733806@hypernet.com> <slrn7q2459.6nk.wtanksle@dolphin.openprojects.net>
Message-ID: <37a7dd2e.1803965294@news.triode.net.au>

On Fri, 30 Jul 1999 02:40:42 GMT, wtanksle at dolphin.openprojects.net (William Tanksley) wrote:

>On Wed, 28 Jul 1999 16:38:24 -0500, Gordon McMillan wrote:
>
>try:
> for x in list:
>   if iffers(x): raise "no!"
>   yadda(x)
>except "no!": pass
>else:
> twiddle()

Not guarranteed to work. Try instead:

exc = "no!"
... raise exc
..
except exc: ..

Python requires the object of a raise to be the same object
as that in the except clause: it is not enough for them to
compare equal as values. Your example only works because
python interns string constants; an implementation detail.

My python interpreter doesn't intern string constants.
The code above will still work, but only because I have extended
the semantics to allow matching by value (which will break
some technically correct python codes).

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From philh at vision25.demon.co.uk  Sat Aug  7 17:55:11 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sat, 07 Aug 99 21:55:11 GMT
Subject: Tkinter -- tarred up documentation?
Message-ID: <934062911snz@vision25.demon.co.uk>

On the Pythonware site (www.pythonware.com) there's a some Tkinter 
documentation entitled "Introduction to Tkinter", and "Tkinter
Class Reference pages". 

Unfortunately these are stored as single HTML pages. Is there anywhere
where they are tarred and gzipped up, so I can easily download them 
to my PC?

-- 
Phil Hunt....philh at vision25.demon.co.uk




From phil at ricochet.net  Mon Aug  9 02:17:23 1999
From: phil at ricochet.net (phil)
Date: Mon, 09 Aug 1999 06:17:23 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com> <37AA804D.9A1317C0@houbits.com> <934115482snz@vision25.demon.co.uk>
Message-ID: <37ae686e.59511102@news.ricochet.net>

On Sun, 08 Aug 99 12:31:22 GMT, philh at vision25.demon.co.uk (Phil Hunt)
wrote:

>In article <37AA804D.9A1317C0 at houbits.com> nhead at houbits.com "Nigel Head" writes:
>> There's no licence statement in the distribution at all for now --
>> my intention is that anyone can use the contents anyway they want,
>> for non-commercial purposes, providing they keep a reference to the
>> original source. Oh yes, and I can't take any responsibility for errors.
>> 
>> Perhaps one of you experienced people can offer some quick advice as to what I
>> need to do -- do I need to just add a licence.txt file or must I add a header
>> to every file? What standard text can I use? Is my intent compatible with the
>> GPL?
This post raised some concerns for me since i have a Python extension
that i was going to add to our latest product release. The product is
not written in Python, nor does it require Python to be running and
the extension is there because i thought it was cool. However, i have
to ask whether it is legal to retain this functionality.
>
>No. The GPL allows anyone to use the software, for commercial or
>non-commercial purposes. However, if they distribute the software, 
>they must include all source code, including all source code they
>write which is linked to the software.
>
I don't understand this. What is "the software"? As far as i can tell,
if the software is distributed in binary format (say as a DLL on
Windoze boxes) then it can be distributed freely with no conditions.
I've written an extension to a vehicle tracking product that we
produce that offers a scripting interface. I'm not sure if anyone will
use this, but i certainly can't distribute it if i have to distribute
the Python source. Added to which, i don't want to distribute my
source (except to selected folk) as i don't need the support grief.

>The license that you want is something like the BSD license except that
>it forbids commercial use.

I can't forbid commercial use. This was why i chose Python. I have no
idea if people can or will  make money on my interface. If Python
imposes commercial limitations then i'll have to drop it, pissed-ly,
since i spent quite a bit of time writing an extension, but that's the
way it goes.

So, can somebody tell me if i can ship the Python Bits and my own DLL
to anybody to use in any way they desire? No fairly otherwise eh?

phil.
>
>-- 
>Phil Hunt....philh at vision25.demon.co.uk
>




From dfan at harmonixmusic.com  Fri Aug 20 09:39:22 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 20 Aug 1999 09:39:22 -0400
Subject: Python to Perl Conversions
References: <1276999045-22443458@hypernet.com> <37bce250.48509376@news.erols.com>
Message-ID: <wk4shud111.fsf@turangalila.harmonixmusic.com>

kernr at mail.ncifcrf.gov (Robert Kern) writes:

| On Fri, 20 Aug 1999 00:27:48 -0500, "Gordon McMillan"
| <gmcm at hypernet.com> wrote:
| 
| >Couple comments -
| >
| >Tom Christiansen wrote:
| >>                                 Yes, slices in python don't include
| >>                                 the termination point.
| >
| >This might be a surprise if you're not used to it, but it turns out 
| >to be very nice. Basically any way of expressing a range will result 
| >in something that includes the left hand side, but excludes the right 
| >hand. So when you're slicing, chopping up, inserting... you (almost) 
| >never have to use "i-1".
| 
| It also means that len(seq[start:end]) == end - start.  It is also
| convenient when using negative indices.

For the record, this is how slicing-style stuff works in the STL in
C++ as well; foo.begin() and foo.end() give you 'first element' and
'one past the end' iterators, and all the generic functions like
for_each take pairs of iterators with that convention.

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From skaller at maxtal.com.au  Sun Aug 15 13:18:05 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sun, 15 Aug 1999 17:18:05 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <m3vham44g5.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <37b6e9c3.345781227@news.triode.net.au>

On 11 Aug 1999 18:23:06 +0100, Michael Hudson <mwh21 at cam.ac.uk> wrote:

>skaller at maxtal.com.au (John (Max) Skaller) writes:

>I'm interested (do you have any code yet?

	Not for the compiler: I'm implementing
the interpreter first. [I have done some work on the
type inference algorithm, to make sure it is
feasible]

>) but I have a few questions:
>
>> The interpreter aims to be compatible with 'pure python':
>> python not using external C modules, and not fiddling
>> implementation details too much. Some of the fiddles
>> will work, and some will not. 
>
>I presume fiddling with sys.exc_traceback.f_back.f_locals is out?

	At present sys.exc_info() will return a tuple in which
the third element is a traceback object. At present, that
traceback object doesn't match the Python one --
it just has a filename and line number in it, to allow the
traceback to be printed.

	However, I will try to make everything
as 'python compatible' as possible, at least in the
interpreter. In the compiler, a few things that
work in the interpreter may fail. However,
another possible result is simply to make the
compiled code more inefficient, rather than it
failing. For example, 'exec' must still work.

	[Interscript uses exec, so I have to support it ;-]

>> Compatibility:
>> 	2) rebinding module variables after importing is not permitted
>
>Does this mean the following is illegal:
>
>import foo
>foo.bar = 1

	Technically yes, according to what I said.
But actually, something less stringent will be required.
What will happen is that the compiler will begin
by executing the interpreter on all modules the main
program imports.

	During that phase of the compilation
process, assignment to module variables outside
the module will be OK.

	AFTER all the modules are imported,
it may not be OK because then I may inline the 
value, which means that an assignemt wouldn't
be honoured. However, it isn't clear that I cannot
simply detect such direct assignments and force
the system to honour the current value,
indeed, this may be necessary in some special
cases such as assigning to 

	sys.stdin

which is a necessary evil for compatibility.

	It may end up that the restriction is
that you cannot do:

	exec "foo.bar = 1"

because that cannot be detected easily by 
static analysis.

>But can foo then have a method set_bar? I guess that would be easier
>to cope with for the analyser.

	Yes. It is OK to mutate objects, as opposed
to rebinding variables: mutation has to be supported.
You may think these things are equivalent. Indeed,
they are in interpreted python. However, the restriction
leads to a performance gain; without some restrictions,
compiled python will not execute faster than the bytecode
interpreter.

	Indeed, the existing python bytecode compiler
makes such restrictions itself [on class __getattr_ 
and __set_attr__ methods, which are cached for
efficiency, as I understand it]

>> 	3) None is a keyword.
>> 	4) range is a keyword
>
>What about other builtins? 

	Yes.

>Can't your static analysis find rebindings
>of builtin names (including range) rather easily? - particularly as
>the don't happen very often.

	I hadn't considered that, but the answer is probably
'yes', provided they don't occur in 'exec's.

>> 	5) What restrictions can you live with?
>
>One thiung I occasionally find very handy is the ability to assign
>methods to classes (or more usually within classes).

	OK. I would like to do what Guido has done in the
optimisation of functions [fast lookup for statically known variables,
plus a dictionary to support others]: provide an optimised structure
where possible, and then provide a dynamic one
for the other cases.  That is, the ideal would be to provide
a 'fallback' to the interpreter.

>> Extensions:
>
>> 	2) optional values in dictionaries (Defaults to None)
>
>Now this one I don't get; do you mean dict[] has dict.get like
>semantics?

	No, you can just write:

	d = {1,2,3}

and it is OK and means:

	d = {1:None, 2:None, 3:None}

>> 	3) 'in' applies to dictionaries 'as if a sequence of keys were used'
>
>Seems unecessary to me. Maybe that's just me.

	The idea is that:

	x in {1,2,3}

means 'x ==1 or x == 2 or x ==3', that is, the construction 

	{1,2,3}

can be used like a set (without introducing a new data structure}.

>> 	5) optionally typed function parameters
>
>Out of curiosity, what syntax do use for this? 

	My first idea was the obvious:

	def f(a:int): ..

or perhaps

	def f(a: IntType): ...

and the interpreter semantics would be 

	if type(a) is not InType: raise TypeError, "Int expected for arg1 of f"

However, in the compiled version, the semantic might be to generate a compiler
error: that is, the following might not work:

	try: f(1.0)
	except TypeError: pass # just skip the call


>I presume if a
>parameter is typed as `Foo' then a value of type `Bar' where Bar
>derives from Foo is allowable.

	I guess that would make sense (the extension isn't
implemented yet -- not a lot of point appealing for comments
_after_ implementing the extension, only to find someone
has a better idea)

>> 	6) what do you want?
>
>Can you derive from basic types? That might make some stuff harder.

	Not at present. Can you give an example where
it is useful? [It can be made to work I think: is it useful?]

>> 	It is difficult to optimise individual
>> modules. It is a different story to optimise
>> a whole program, where _every_ call
>> to a function can be traced. Of course,
>> this may involve restricting what 'exec' can do,
>
>Like banning it entirely, I suspect.

	No. I cannot do that because the program
that I want to compile, interscript, uses exec,
and depends on it to execute client script.
This is the major feature of interscript, so exec
has to work (at least with restrictions).

	That means the compiled code has to
contain a full run time system (unless it can be
detected that it isn't required).

>> and it may involve other restrictions
>> (such as not  changing any module bindings
>> after loading).
>
>Is that really that much of an issue? I'd have thought that they'd be
>fairly easy to spot. 

	Several people have said that, and may be right.
I'll examine this issue, since it would provide extra compatibility
to permit it.

	As an example of why rebinding should be supported:

	_open = open
	opened_files = []
	def open(f,m): 
		opened_files.append(f)
		_open(f,m)

This code just hooks 'open' to create a list
of all files opened, which could be useful.
So rebindings are a useful trick.

>Mind you, the whole thing sounds sooo hard, that
>anythong that makes it easier is good.

	The things that turn out to be hard are
often surprising. For example, right now I having
major problems with something that sounds trivial:
getting the interpreter to set the line number
so that a traceback can print out as it does in Python.

	That sounds easy, but the LR grammar I'm
using is not well adapted to it -- NEWLINE tokens
come after statements, instead of before them,
where the line number is needed.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From I.Clarke at strs.co.uk  Tue Aug  3 08:44:33 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Tue, 03 Aug 1999 13:44:33 +0100
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk>
Message-ID: <37A6E431.23BF98F1@strs.co.uk>

Just browsing through the Python documentation, I also notice that it is
possible to restrict the execution of programs, similar to Java's sand
box (called a "padded cell" in Python-speak).  This opens up a very
interesting possibility - namely that if the system I described in my
previous post allowed CGI scripts to run in a restricted environment,
then there would be no reason why an ISP could not allow their users to
use CGI in their webpages (at present, most ISPs won't grant CGI access
to their users due to the security nightmare that would present).

I intend to do some work on this as soon as I get time.  If anyone wants
to help please email me.

Ian.



From tim.jarman at lineone.net  Fri Aug 13 19:47:52 1999
From: tim.jarman at lineone.net (Tim Jarman)
Date: Fri, 13 Aug 1999 23:47:52 +0000
Subject: Glade for the Mac
References: <199908131425.KAA14896@python.org>
Message-ID: <37B4AEA6.CD0C67AE@lineone.net>

Hello all
I have been following the recent discussion on Python GUI tools with great interest - Glade sounds like the answer to (many of) my prayers. I develop on, though not just for, the
Mac, so imagine my joy on discovering a Mac version of Glad at the Glade home page.

Alas I don't have CodeWarrior. Could anyone either:

(a) point me to precompiled binaries (I only need PPC) or
(b) mail me ditto (if not too huge) or
(c) suggest how I could compile the source under MPW?

Many thanks

Tim Jarman
Britannia Consulting Ltd.
"A smile, a song, and a compiler"




From thantos at brimstone.mecha  Fri Aug  6 19:15:45 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Fri, 06 Aug 1999 23:15:45 GMT
Subject: I need some advice on python
References: <7oepao$6hn$1@nnrp1.deja.com>
Message-ID: <slrn7qmqso.igt.thantos@brimstone.mecha>

On Fri, 06 Aug 1999 13:52:28 GMT, timcollinsdeja at my-deja.com
<timcollinsdeja at my-deja.com> wrote:

>I've seen a couple interesting projects using python for this type of
>stuff, but they either run on Unix (Know-bots) or OS/2 (Rover).
>Anything out there for Windows?

There's no really good reason that the Python agent code floating
around /can't/ run on Windows; most of the networking code will work
regardless unless they're doing really deeply fancy things.  If they
are, you can very likely turn one of the Distributed Object harnesses
written in Python to your use as a means of migration.  One of the
advantages of doing so is that agent docks needn't be limited to
/just/ Windows or Unix but can span both transparently if you're
shipping Python bytecode around the network.

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From fredrik at pythonware.com  Fri Aug 20 04:42:33 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 08:42:33 GMT
Subject: More random python observations from a perl programmer 
References: <37BC6986.A66B044E@starmedia.net> <199908191634.KAA01333@jhereg.perl.com>
Message-ID: <012d01beeae7$f3646320$f29b12c2@secret.pythonware.com>

as so often before, Tom ran out of positive energy and
started posting stuff like:

> You seem to be looking for a fight.   Please say nothing
> at all if that's all you want to do.

> Fine.  I just hate being told to read the fucking manual when there
> ISN'T one.  I despite Linux and GNU and the FSF and all that crap
> because of this.

> The fact that they did icky windows help and not man pages says
> something significant.

> --tom

some days, this guy really works hard to live up
to his first name:

    Svenskt uppslagsord:
        tom tomt tomma (adj.)
        
        without content (also fig. "lacking substance,
        meaning, or value")

    English translation:

        empty

> Is your newsreader broken, or haven't you ever bothered to
> read news.announce.newusers to learn how to use Usenet?

comp.lang.python is not your ordinary usenet.  please read
the comp.lang.python posting guidelines before you post more
crap like this.  this is not comp.lang.perl.misc.

</F>




From philh at vision25.demon.co.uk  Tue Aug 31 17:26:51 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 31 Aug 99 21:26:51 GMT
Subject: Parrot: Why the clunky syntax?
References: <935279776snz@vision25.demon.co.uk> <935454399snz@vision25.demon.co.uk> <37C2B9CE.95AA5FEC@lri.fr> <37C9FF53.DD466783@compaq.com> <19990830150321.C6858@optiplex.palga.uucp> <7qg7ek$k5e$1@newshost.accu.uu.nl>
Message-ID: <936134811snz@vision25.demon.co.uk>

In article <7qg7ek$k5e$1 at newshost.accu.uu.nl>
           m.faassen at vet.uu.nl "Martijn Faassen" writes:
> So what Greg was asking, why not use a more Pythonesque syntax for parrot,
> where indentation has actual meaning?

1. I like the current syntax

2. I have already written the parser (although i will enhance it to
do better error handling -- that's on the TODO list)

3. It wouldn't be particularly difficult to add a more Python-like
front end; I might do this sometime if I am feeling bored and have 
nothing better to do; at the moment writing a tkinter backend is 
the highest priority for Parrot, then it will actually be useful

-- 
Phil Hunt....philh at vision25.demon.co.uk




From jeremy at cnri.reston.va.us  Thu Aug 19 13:14:09 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 19 Aug 1999 13:14:09 -0400 (EDT)
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc29d8@cs.colorado.edu>
References: <37bc1339@cs.colorado.edu>
	<slrn7ro8n1.h6v.jblaine@shell2.shore.net>
	<37bc29d8@cs.colorado.edu>
Message-ID: <14268.15201.81468.481452@bitdiddle.cnri.reston.va.us>

[Tom Christiansen <tchrist at mox.perl.com> writes:]
>Well, it's the polymorphism thing.  If "+" is supposed to work on anything
>(which I don't agree with, but that's not my decision), I don't understand
>why "in" shouldn't.  I keep thinking that everything should work on

I kind of agree with you.  I often use in with a dictionary without
thinking about it and then go back to fix it.  (I'm quite happy that +
works on many types BTW.)

It is interesting to note that early version of JPython implemented in
for dictionaries, based on exactly the reasoning you describe.  It
ended up leading to some puzzling behavior when used in a for loop.
I'm hazy on the detail, but I think the exact behavior was: JPython
would treat a dictionary as a sequence within a loop context provided
that there was a 0 key (plus zero or more successive integers).  I
passed a dictionary without a 0 key; JPython caught an exception and
treated the dictionary as an empty sequence.  I hadn't intended to use 
a dictionary in the for loop, rather a function had mistakenly
returned a dictionary rather than the dictionary's keys.  

I'm not sure what the moral of the story is, except that JPython
doesn't implement in for dictionaries anymore and I'm mostly happy
about that.

>everything.  For example, this surprises me:
>
>    x = {1:2, 3:4}
>    y = {10:20, 30:40}
>    z = x + y
>
>The latter isn't legal.  But in Perl, you can merge hashes easily.
>
>    %x = (1,2,3,4);
>    %y = (10,20,30,40);
>    %z = (%x, %y);
>
>    print $z{1}
>    2
>    print $z{30}
>    40
>

You can do almost exactly this in Python using the update method on
dictionaries.  

    x = {1:2, 3:4}
    y = {10:20, 30:40}
    z = {}
    z.update(x)
    z.update(y)
    print z
    {30: 40, 3: 4, 10: 20, 1: 2}

As you noted with the [].sort operation, you need to explicitly create 
the new object that call some methods on it.  In my book it is at
worst a minor annoyance.

>:Thanks for the document.  It was very interesting reading.
>
>You're welcome.  I got so fed up with the damned flames coming
>from people who don't actually even know what they're talking about,
>that I decided to make sure I wasn't one of them. :-)
>

Indeed, I enjoyed reading both your posts this morning.  While I don't 
agree with all of your criticisms, I think you make many good points.
I did notice that some of your criticisms were out-of-date, perhaps
reflecting Python previous to 1.5.2.  (And some of your criticisms are 
already being addressed, such as the string methods that are in the
current CVS tree but not yet release.
    x = 'asdbasdas'
    x.split('s')
    ['a', 'dba', 'da', '']
    # instead of:
    # import string
    # string.split(x, 's')
)

Jeremy



From aeritargyl at aol.com  Mon Aug 16 01:29:43 1999
From: aeritargyl at aol.com (aeritargyl at aol.com)
Date: Sun, 15 Aug 1999 23:29:43 -0600
Subject: ONLY ADULTS 61427
Message-ID: <150899232943@aol.com>

Free XXX:

http://207.240.225.250






h,ERUe4+



From fw at s.netic.de  Fri Aug 20 07:06:34 1999
From: fw at s.netic.de (Florian Weimer)
Date: 20 Aug 1999 13:06:34 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <877lmqaeyt.fsf@deneb.cygnus.qad.org>

Tom Christiansen <tchrist at mox.perl.com> writes:

> GOTCHA: (low)
>     Regexes default to emacs style, not egrep style!  Gads!  This is
>     surprising to anyone except an emacs weenie.  Sigh.  And just *try*
>     to read the manpage on the differences based on passed in arguments
>     establishing the style.  There aren't any manpages!  Have a nice day.

That's no longer true.  There's now a regular expression module
in the standard distribution based on the PCRE library which uses
Perl-compatible regular expressions (with some very helpful extensions,
like named groups).



From philh at vision25.demon.co.uk  Fri Aug 13 17:48:50 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Fri, 13 Aug 99 21:48:50 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov>
Message-ID: <934580930snz@vision25.demon.co.uk>

In article <37B47506.14EBD269 at msfc.nasa.gov>
           drew.mcdowell at msfc.nasa.gov "Drew McDowell" writes:
> I looked in the FAQ and couldn't find anything on the Compound
> Assignment Operators.
> (+=, *=, -=, etc..)
> Why dosen't Python support them?

Cos Guido doesn't like them, I imagine.

> Is there a easy way to add them?

Well, you have the source...

-- 
Phil Hunt....philh at vision25.demon.co.uk




From robert_j_roberts at rl.gov  Sun Aug 29 21:27:46 1999
From: robert_j_roberts at rl.gov (Roberts, Robert J)
Date: Sun, 29 Aug 1999 18:27:46 -0700
Subject: fo.flush() & fo.write() on windows
References: <000001bef280$597f4040$a22d153f@tim>
Message-ID: <37C9DE12.7498090E@rl.gov>


Tim Peters wrote:

> [Roberts, Robert J]
> > Am I asking too much?
>
> On Windows, yes.  C's stdio isn't its native file-handling system, and its
> implementation of stdio is, umm, "surprising".
>
> > pseudo code:
>
> >     fo = __builtin__.open(foName, 'a')
>
> Why not plain "open"?

Because __builtin__.open() works? I started doing file opens by copying some of
Guido's code. It worked. I never tried open() alone.

>
>
> >     foSize = os.stat(foName)[6]     # if foSize = 100...
>
> os.path.getsize(foName) is easier on the brain (new in 1.5.2 though).

Thanks. Didn't know it was an option.

>
>
> >     fo.write('Some string.')             # fo.write() returns None
>
> file.write() always returns None, as documented.  Don't know what you expected
> it to return, but read the docs.

I must have been reading the wrong place in the documentation all this time. I was
reading doc/lib/node105.html which says:

"write(fd,str)  Write the string str to file descriptor fd. Return the number of
bytes actually written."

>
>
> >     foSize = os.stat(foName)[6]    # ...foSize STILL = 100.
>
> Unsurprising so far.
>
> >     fo.flush()
> >     foSize = os.stat(foName)[6]    # ...foSize STILL = 100.
>
> Ah -- *that's* Windows <wink>!
>
> >     fo.close()                     # 'Some string' appears in fo.
> >     foSize = os.stat(foName)[6]    # [now the bytes appear]
> >
> > I am running Python 1.5.2 on Windows 95. fo.write() appears to be
> > buffering the write and returning None for bytes written on
> > fo.write().
>
> "None" is expected (see above).  Buffering should be expected too since you
> opened the file with buffering.  If you don't want buffering, pass 0 as the
> third arg:
>
>      fo = open(foName, 'a', 0)

Didn't know this was an option. I'll give it a try and see what it does. But (as
you say), it probably won't do much if Windows is still buffering.

>
>
> Alas, on Windows that won't make much of a difference in what you see.
>
> > fo.flush() appears to do nothing.
>
> MS's implementation of stdio's fflush (which Python's file.flush() calls
> directly) does clear the stdio buffers, but merely flushes them to Windows'
> internal buffers.  It does not imply commitment to disk.  The only way to get
> that under Windows is to pass the non-standard "c" (for Commit) mode flag to
> "open":
>
>     fo = open(foName, 'ac')

I'll look into this too, but I don't want to slow the code down anymore if I can
help it. I'll compare it with closing and reopening the file (which DOES dump the
buffer immediately) just to see what happens.

>
>
> Then fo.write(whatever) *still* won't (necessarily) show up on disk right away,
> but fo.flush() will work the way you expect.  "c" is unportable, and much as
> you may think you want it now <wink>, slows down Windows output enormously.
>
> windows-was-designed-under-the-assumption-that-it-would-
>     never-crash<snort>-ly y'rs  - tim

Thanks, millions!




From jstevens at basho.fc.hp.com  Fri Aug 13 23:06:26 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:06:26 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132005.OAA03204@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:05 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cf22@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :Ok.  I'll take that challenge. . . create for me an unreadable
> :Python program that actually compiles and runs.
> 
> Ok, I just showed the paperboy some python programs, and he didn't 
> think were readable at all.

"ME".  The statement was: "ME".

Take that same paper boy, train him in Perl and Python at the same
time, and see which one he prefers.  Go on, I'll wait. ;-)

> Please face it: terms like "readable" and "intuitive" are completely
> subjective, and therefore purely opinion.

This is incorrect.  They are not subjective.  They can be measured.

"Intuitive", when meant as: "leveraging off of existing learning
and habits, combined with techniques that take into account human
perceptual patterns" is both quantifiable and quite objective.

Visual acuity is, in fact, built right in to the human eye/brain.
It's not even cultural.

Since Python enforces visually acute coding practices, in general
Python will be more readable (since, whether you admit it or not,
Perl'ers span the range from using no whitespace at all, to using
wildly different mixes of indentation styles).

John S.




From gmcm at hypernet.com  Wed Aug  4 14:45:37 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 4 Aug 1999 13:45:37 -0500
Subject: Win32 Installer - new beta
Message-ID: <1278333954-16187909@hypernet.com>

 A new beta (02) of the Win32 Installer package is available on
starship:

 http://starship.python.net/crew/gmcm/install.html

This release incorporates bug fixes and some enhancements:

  The major changes are that command line arguments are 
  passed into your script, and a standalone / installer can be run 
  from anywhere -it is no longer assumed that the current working 
  directory is also the directory the program lives in. Files are 
  always unpacked into the "home" directory, and the absolute path of 
  the "home" directory is always available. 

  See http://starship.python.net/crew/gmcm/install.html#beta02 for 
  details.

What is it?
 Briefly, it is a sort of compilerless Freeze - a way of distributing
Python apps with all the required Python support included.

 Python-style license.

Contact: gmcm at hypernet.com


- Gordon



From msteed at altiris.com  Thu Aug 19 13:46:04 1999
From: msteed at altiris.com (Mike Steed)
Date: Thu, 19 Aug 1999 11:46:04 -0600
Subject: How to sell Python to your boss?
Message-ID: <51CBA4F0B84CD211BBA60060080588D27ABC61@exchange.keylabs.com>

My take would be:

a) Argue that you could quickly develop a prototype in Python and then
rewrite it in C once you have it working (although in reality you have no
intention of doing this).

b) Once it is done, you have a decent chance of convincing him that the
maintenance will be much easier in Python -- including the time for other
developers to learn the language -- if you compare some of the clean, easy
Python code to its equivalent C ugliness (sounds like string parsing is one
good candidate).

And then buy him a copy of _Learning Python_ to give to the future
maintainers.  They won't be sorry.

-Mike

-----Original Message-----
From: Petri Mikael Kuittinen [mailto:eye at niksula.hut.fi]
Sent: Thursday, August 19, 1999 10:54 AM
To: python-list at python.org
Subject: How to sell Python to your boss?


I proposed an interesting idea to my employer for my final thesis.

My proposed final thesis includes:
- a prototype implementation of networked multi-player video gambling
game (which also has a single-player mode)
- a design of a network protocol to allow this and automated testing
of the game (including several automated test player programs which
have different kind of AI). This protocol and test system is designed
so that it can easily be expanded to handle any sort of gambling game

Most of the work is design, documentation and programming a
rudimentary, but easy-to-use, GUI for the multi-player game. My boss
otherwise agreed with my idea ("it would be really useful..."), but he
doesn't want me to do the prototype implementation using Python. He
said I must use a program language that all the other programmers in
our R&D department understand. Well, that leaves only one choice: C

[snip]



From sholden at pgrad.cs.usyd.edu.au  Fri Aug 13 23:36:23 1999
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 14 Aug 1999 03:36:23 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ce93@cs.colorado.edu> <slrn7r9la9.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nde.h3c.jstevens@bamboo.verinet.com>
Message-ID: <slrn7r9p2h.pu8.sholden@pgrad.cs.usyd.edu.au>

On Sat, 14 Aug 1999 03:08:33 GMT,
    John Stevens <jstevens at bamboo.verinet.com> wrote:
>On 14 Aug 1999 02:32:12 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>>On 13 Aug 1999 20:04:03 -0700,
    John W. Stevens <jstevens at basho.fc.hp.com> wrote:
>>>
>>>Perl doesn't have lists.  Python doesn't have built-in arrays.
>>
>>You should learn some perl you now..
>>
>>@array = (1,10,20,30);
>>$from_list = (1,10,20,30);
>>$from_array = @array;
>>print "$from_list\n$from_array\n";
>>
>>Will output :
>>30
>>4
>
>The @ prefix denotes an array.  You, yourself, should learn
>Perl.  Calling an array a list, doesn't make it one.

Can you read?

Can you see a @ in the following line of code :

$from_list = (1,10,20,30);

No you can't.  Is there an array in that line of code? No. Is there a list
in that line of code? Yes. Do lists and arrays behave differently? Yes, just
look at the different outputs.

>
>>Perl has lists,
>
>Not built in, it doesn't, unless you define array and list as being
>different words for exactly the same type/class.

I don't think you can get more built in then perl lists.

Again I repeat, here is a perl list : ('a', 'b', 'c') or qw(a b c)

>
>>if you know perl you would know this.
>
>I know Perl.  You need to learn Python.

Did I mention a single bit of python syntax in my post? No. 
Was I discussing python syntax? No.
Do you know if I use python? No.
Do you know if I like python? No.
Do you know if I have already spent time learning python?
Do you know if I am currently learning python?

What was the point of saying I need to learn Python? I wasn't saying python
was worse than perl, I wasn't saying python syntax is strange, I wasn't
talking about python. You might as well have told me to learn C or Lisp.

At least Lisp would be more relevant than python, if you were trying to
tell me to learn about lists in some strange abstract way.

Have I questioned your python knowledge? No.
Have I questioned your perl knowledeg? Yes, but only because you have
consistantly made basic mistakes when speaking about perl.

>
>>If you program in perl 
>>and don't know this, then you must get very very confused at times.
>
>If @ denotes list, then the following Perl would be illegal:
>
>@ary = (1, 2, 3);
>@ary[5] = "Test";
>
>But, obviously, this is not illegal.

@ary[5] is an array slice. It is an array. It happens to be an array 
that constists of only one element. Why would it be illegal? It is just
the special case of :

@ary[1,2,3,4,5,10,20]

It is an array. Thus it has a @.

Used like that it is almost a sure indication of an error by the programmer,
because a single element array slice has no benefit over the simple
array element in that context.

>
>>>I will assume that a list module is available for Perl.
>>
>>No it is one of the built in bits... like hashes and arrays.
>
>Really?  What is the prefix character that denotes a list?

You can't have a list in a variable. Lists in perl are constant. So
what would be the use of putting them in a _variable_. Just because there
is no list variable does not mean there are no lists. Again here is a list :

(1,2,3,4,5,6,7,8)

>
>>>I wasn't trying to compare features, I was simply pointing out
>>>that your comparison was Apples and Oranges, and therefore at
>>>least somewhat invalid.
>>
>>Only because you have no idea what you are talking about.
>
>:-)
>
>Coming from somebody who doesn't know the difference from *EMULATING*
>a list with an array, vs. a real array, that is a good one!

Again here is a list :

(1,2,3,4,5,6,7,8)

That is not an array. It is a list. It is different from an array. I already
showed how assigning to a scalar gives different results for lists and arrays,
here are some more differences :

push( (1,2,3,4,5), 6); 

This results in an error message, funnily enough the message is :
Type of arg 1 to push must be an array (not list)

Still claim perl has no lists? Of course you know better then perl itself.

@array = (1,2,3,4,5);
push(@array,6);

That works fine, because we are passing an array to push not a list.

Pop is the same :

pop (1,2,3,4,5);

Is an error (the message is as above s/push/pop/)

@array = (1,2,3,4,5);
pop @array;

Is legal and pops the 5 off @array.

>
>Now I suppose that you will tell me that Perl has stacks, too!
>;->

No it doesn't. 

-- 
Sam

why can't newbies use hash slices in their hello world programs? :-)
	-- Uri Guttman in <x74skxhve5.fsf at home.sysarch.com>



From gutier at intergate.bc.ca  Sun Aug 22 23:49:09 1999
From: gutier at intergate.bc.ca (Gerald Gutierrez)
Date: 22 Aug 1999 20:49:09 -0700
Subject: Python / Cool looking applications?
Message-ID: <m3ogfz9mx6.fsf@intergate.bc.ca>

Hi all.

I've been looking into using Python as UI glue and at Python's imaging
extensions. Are there any examples of very cool looking interfaces or
otherwise well-designed interfaces that are built in Python?

Also, does anyone know of any applications that use Python's imaging
extensions?

Thanks very much.

Please forward replies to gutier at intergate.bc.ca.





From claird at starbase.neosoft.com  Mon Aug 23 09:39:15 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Aug 1999 13:39:15 GMT
Subject: syntax
References: <1278776075-25895447@hypernet.com>
Message-ID: <5B279F40BA8B81C4.99FEFD50B6F45FD8.44C803E9F3EA1297@lp.airnews.net>

In article <1278776075-25895447 at hypernet.com>,
Gordon McMillan <gmcm at hypernet.com> wrote:
>Xavier Bec writes:
>> 
>> Why are the double dots ":" necessary after a if, for or while
>> instruction ?
>> All the others instructions are terminated with a cariage return but
>> not condition or loop ones. I know that a semi-column is facultative
>> and only used if you have to put a second instruction on the same
>> line. Why is not the same for if, etc.
>
>From the standpoint of parsing, the colon is redundant to the 
>following INDENT. As I understand the history, their usage is 
>mandatory because of user testing done on ABC, which was one of a 
>host of influences on the design of Python.
>
>For some reason, the dominant hominid specie of planet Earth is 
>uncomfortable without a certain level of redundancy. Some members 
>thereof have taken this predeliction to ridiculous extremes, creating 
>languages such as "Java".
			.
			.
			.
Fredrik labeled the controlled dimension "readability".
As I understand the description, it specifically has to
do with accessibility for *beginners* in a language.
This inspires two reactions:
1.  We need a word to distinction the quality
    that people-who-don't-already-know-it-can-
    read-it.
2.  Maybe this'll help sharpen the sniping
    between proponents of Python and Java
    syntaxes about how the latter is slanted
    *too* much toward newbies.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From eye at niksula.hut.fi  Thu Aug 19 12:53:38 1999
From: eye at niksula.hut.fi (Petri Mikael Kuittinen)
Date: 19 Aug 1999 19:53:38 +0300
Subject: How to sell Python to your boss?
Message-ID: <ka8yaf7d84t.fsf@delightful.cs.hut.fi>

I proposed an interesting idea to my employer for my final thesis.

My proposed final thesis includes:
- a prototype implementation of networked multi-player video gambling
game (which also has a single-player mode)
- a design of a network protocol to allow this and automated testing
of the game (including several automated test player programs which
have different kind of AI). This protocol and test system is designed
so that it can easily be expanded to handle any sort of gambling game

Most of the work is design, documentation and programming a
rudimentary, but easy-to-use, GUI for the multi-player game. My boss
otherwise agreed with my idea ("it would be really useful..."), but he
doesn't want me to do the prototype implementation using Python. He
said I must use a program language that all the other programmers in
our R&D department understand. Well, that leaves only one choice: C

Some of my game programmer colleagues know a little bit of C++, Java
and Perl, but these languages cannot be used either. Operating system
or CPU power is not a problem here (I can use Linux and several
hundred MHz of CPU power).

I have lots of programming experience in C, and much less in
Python. Despite this I am confident that I would be able to implement
the prototype using Python many times faster than using C. Sure I can
do it with C, or even with assembler (I have made a GUI with m68k
asm), but parsing text strings and prototype GUI design is so
laborious in C, that I would only be able to implement a tiny portion
of the whole thing.

I can understand my bosses opinion. He wants to protect the interest
of this firm. If I would leave the firm, then someone else would have
tp spend time learning Python in order to modify my program code.

For several months I have tried to get my colleagues excited to learn
Python, but so far nobody has been interested to learn new languages,
except for some mild interest towards Java (which is IMHO an
overly-hyped language).


Petri

-- 
<(O)> Petri Kuittinen, also known as Eye, Dj Eye or Peku               <(O)>
<(O)> ADDRESS: Timpurinkuja 1 C 39, FIN-02600 Espoo, Finland           <(O)>
<(O)> EMAIL: eye at iki.fi WWW: http://www.iki.fi/~eye/ PHONE: 09-5472380 <(O)>
~The only way to avoid mistakes is to do nothing or try not anything new.



From mwh21 at cam.ac.uk  Thu Aug 19 18:50:44 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Aug 1999 23:50:44 +0100
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu> <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk> <37bc87fa@cs.colorado.edu>
Message-ID: <m3lnb7e663.fsf@atrus.jesus.cam.ac.uk>

Tom Christiansen <tchrist at mox.perl.com> writes:

>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.python, 
>     Michael Hudson <mwh21 at cam.ac.uk> writes:
> :Ah, have you discovered rlcompleter yet? Put this in your ~/.pythonrc:
> :
> :import rlcompleter

Agh! Duh!

Try making that 

import rlcompleter, readline

(sound of forehead being smacked)

> :readline.parse_and_bind('tab: complete')
> :
> :set PYTHONSTARTUP=~/.pythonrc and then in python type:
> :
> :d={}
> :d.
> :
> :and whack tab a couple of times. This should definitely be better
> :advertised; I use it all the time.
> 
> That just triggers:
> 
>     % python
>     Python 1.5.1 (#1, Sep  3 1998, 22:51:17)  [GCC 2.7.2.3] on linux-i386
>     Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>     Traceback (innermost last):
>       File "/home/tchrist/.pythonrc", line 2, in ?
> 	readline.parse_and_bind('tab: complete')
>     NameError: readline
> 
> Time to do a real install.    Hate redHate.

You might need to do that too.

HTH
Michael



From jhspies at futurenet.co.za  Fri Aug 13 15:59:20 1999
From: jhspies at futurenet.co.za (Johann Spies)
Date: Fri, 13 Aug 1999 19:59:20 GMT
Subject: New Gadfly fails test
Message-ID: <Pine.LNX.3.96.990813215602.427A-100000@Johann>

Hallo

I tried the recently announced kwPtest.tgz. 

After unpacking it in /tmp, running python gfinstall.py and python
gftest.py the test ended in 

select * from lazy
Traceback (innermost last):
  File "gftest.py", line 723, in ?
    retest(directory)
  File "gftest.py", line 702, in retest
    cursor.execute(s)
  File "gadfly.py", line 355, in execute
    results[i] = cs[i].eval(params)
  File "sqlsem.py", line 2819, in eval
    (tups, attorder) = select_list.map(substseq)
  File "sqlsem.py", line 1286, in map
    values.append(exp.value(assnlist))
  File "sqlsem.py", line 1145, in value
    result[i] = contexts[i][ra]
KeyError: ('LAZY', None)

Is this a bug in the program or on my system.  The previous version of
gadfly (1.0) is working without problem .

Johann

 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "Watch ye therefore, and pray always, that ye may be 
      accounted worthy to escape all these things that shall
      come to pass, and to stand before the Son of man."    
                               Luke 21:36 




From ScherBi at BAM.com  Thu Aug 19 11:08:17 1999
From: ScherBi at BAM.com (ScherBi at BAM.com)
Date: Thu, 19 Aug 1999 11:08:17 -0400
Subject: NEWBIE: Setting WinNT python to point to my tcl DLL
Message-ID: <3F8BA119B48AD11197470008C7A493CB98B513@HQOBGEX1>

find the requested dll, and add the path to it to your path.  ( i hope that
reads better than it looks 8-) )
Using the System applet in control panel, you can modify the path for the
current logged on user, or for everyone if you're logged in as
administrator.

You'll have to exit and restart any dos boxes.  existing ones wont see the
change.

Bill

-----Original Message-----
From: Ron A. Zajac [mailto:zajac at nortelnetworks.com]
Sent: Thursday, August 19, 1999 10:43 AM
To: python-list at python.org
Subject: NEWBIE: Setting WinNT python to point to my tcl DLL



Dear Python Anciens,

  I've installed the latest Python on my WinNT system, but declined the
Tk install, since I already have tcl/Tk8.1.1 on my system.  It appears
to be the case I need to point the system to my current tcl/Tk install,
however.  "IDLE" can't find "tcl80.dll" which, of course, doesn't exist.

  Where can I make this change so IDLE can find the relevant DLL?




-- 
http://www.python.org/mailman/listinfo/python-list



From robin at alldunn.com  Thu Aug  5 00:53:58 1999
From: robin at alldunn.com (Robin Dunn)
Date: Wed, 4 Aug 1999 21:53:58 -0700
Subject: Stupid question about .EXE
References: <7o93uk$sjrn$1@reader2.wxs.nl> <7ob01f$ghn$1@nnrp1.deja.com>
Message-ID: <GN8q3.543$L73.32637041@news.randori.com>

>
> Don't know what the name of the utility is, or if it's executable
> itself or is c source code to allow the making of a python .EXE
>

Check out the freeze utility included with Python.  This requires a C
compiler to build the actual executable, but the utility is usable on all
platforms.

Also take a look at Grodon McMillan's Installer utility found at
http://starship.python.net/crew/gmcm/install.html.  This one doesn't require
a C compiler but is Win32 only (and works quite good in my opinion.)


--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!






From skaller at maxtal.com.au  Fri Aug 20 18:47:42 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Fri, 20 Aug 1999 22:47:42 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au> <7p7h3p$jto$1@newshost.accu.uu.nl> <slrn7rekon.r7j.wtanksle@dolphin.openprojects.net>
Message-ID: <37bdda51.800579923@news.triode.net.au>

On Sun, 15 Aug 1999 23:54:00 GMT, wtanksle at dolphin.openprojects.net (William Tanksley) wrote:


>Best of all, Pop-11 is finally free.

	URL?

>>Hm, why Ocaml? 

>ML does type inferance.  He's probably using that just so  he gets the
>inference free.

	I thought of that, but it probably won't work, at least
not completely; I will be implementing a Python specific
inference engine.
John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From dwalker at iximd.com  Sun Aug 15 12:51:08 1999
From: dwalker at iximd.com (Damond Walker)
Date: Sun, 15 Aug 1999 16:51:08 GMT
Subject: Question: copy/deepcopy with the curses module?
Message-ID: <0eCt3.1775$kW2.100172@typhoon2.gnilink.net>

    I'm writing a small program which uses the standard curses module.
While the program runs I show message boxes, sometimes two levels deep.  At
this point I have a nice layered display.  Good so far.  :)

    Now, the user interacts with the topmost message box so that it
dissapears.  To repaint the screen I'm clearing it and repainting everything
from scratch.  The end result is what I want but the performance is, shall
we say, somewhat pitiful.  On a fast machine the user probably wouldn't
notice but on a slower machine the user can actually track the progress of
the paint process.

    The code is layered like this...

    class SIO: <contains methods which interact with curses>
        self.stdscr = curses.init...()
        self.s = newwin(...)

    class MsgBox(DialogContainer):
        def __init__(self, Screen, Message):
            self.SetProperty("SCREEN", Screen)   # Screen points to instance
of SIO


    All my classes use the same instance of SIO (which basically provides a
wrapper around displaying text on the console).

    What I want to do is get a snapshot of the current SIO instance.  This
would allow me to quickly save the current SIO instance in a temp. variable,
do some screen updates, and then quickly repaint the screen by restoring the
SIO instance to it's older value.

    I tried to use the Copy module (copy and deepcopy) but it complained
about how it couldn't copy the curses stuff.

    Anyone have any pointers? (note: I'd like to do this using the standard
modules which are distributed with Python 1.5.2)


                    Damond





From englim at pc.jaring.my  Fri Aug 20 05:33:44 1999
From: englim at pc.jaring.my (Thomas Lim)
Date: Fri, 20 Aug 1999 09:33:44 +0000
Subject: Reading from serial port and dumping to console
Message-ID: <37BD20F8.9A984D7E@pc.jaring.my>

I'm trying write a program to read from the serial port (barcode reader
attached) and dump to the console, so that my application program can
receive the data.
OS: Red Hat Linux 6
Python: 1.5.2

Program listing:
p = open('/dev/ttyS0', 'r')
t = open('/dev/tty0', 'w')
while 1:
    x = p.readline()
    print x   # yes, I could read from /dev/ttyS0
    t.write(x)  # but this has no effect on console


Hope that someone could help.

Regards,
Thomas.






From skip at mojam.com  Thu Aug 26 11:59:18 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 10:59:18 -0500 (CDT)
Subject: Documentation string
In-Reply-To: <7q3c88$k8m$1@nnrp1.deja.com>
References: <7q3c88$k8m$1@nnrp1.deja.com>
Message-ID: <14277.25545.616066.277845@dolphin.mojam.com>

    sp00fD> Can someone tell me why the
    sp00fD> def foo(self):
    sp00fD> """A method"""

    sp00fD> won't work?

You need to indent it.  The def statement opens a new block, which is
required to be indented further than the def's indentation:

    def foo(self):
	"""a method"""
	if (self.spam):
	    return self.eggs
	return self.spamspameggsandspam

You may be confusing the doc string with a comment.  The following is valid:

    def foo(self):
    # a method
	if (self.spam):
	    return self.eggs
	return self.spamspameggsandspam

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From kaz at ashi.FootPrints.net  Mon Aug 16 01:27:51 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Mon, 16 Aug 1999 05:27:51 GMT
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net> <slrn7re808.77t.kaz@ashi.FootPrints.net> <slrn7refg6.d4.bmcd@ripley.sulaco.net>
Message-ID: <slrn7rf8bf.hp6.kaz@ashi.FootPrints.net>

On 15 Aug 1999 22:18:59 GMT, Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
>There is nothing wrong with i = i + 1 (especially for us Pythonistas :-), but
>taken together with the loop it suggests an unfamiliarity with the accepted
>idioms of the language.

It does suggest that. But it could also be some religous thing. As in ``I won't
use this evil ++ thing because KRUDBAL didn't have it, and KRUDBAL is the
finest programming language there ever was. So if it's not in KRUDBAL, you
don't need it.''

:)



From mithy at my-deja.com  Thu Aug  5 19:32:06 1999
From: mithy at my-deja.com (Preston Landers)
Date: Thu, 05 Aug 1999 23:32:06 GMT
Subject: downloading web based email
References: <7och4d$jr0$1@nnrp1.deja.com> <14249.55454.471036.787883@dolphin.mojam.com>
Message-ID: <7od6tm$4o7$1@nnrp1.deja.com>

In article <14249.55454.471036.787883 at dolphin.mojam.com>,
  skip at mojam.com (Skip Montanaro) wrote:
>
>     Preston> I'm thinking of writing a Python program to download
email from
>     Preston> web-based services such as Yahoo mail, Deja, Hotmail, and
so
>     Preston> on, and deliever the mail in a format that standard unix
>     Preston> mailers can use.

> Presuming the services you're interested in support a standard mail
transfer
> protocol like POP or IMAP, you need look no further than Eric
Raymond's
> excellent fetchmail program:

I use fetchmail on a daily basis and I'd be overjoyed if it supported
these services.  It doesn't, because they don't use standard IMAP or
POP services.  They use only their HTML interface which is different for
each service.  They want you to look at their banner ads.  Have you ever
used a service like this?  Believe me, I'd use fetchmail if I could.

I've never looked at fetchmail's source, but I believe it to be in C,
and designed to handle a specific class of RFC compliant standard mail
systems.  It would take a major reengineering to make it work like I
want, and I don't feel like doing it in C.  So, unless anybody has any
other suggestions, I'll probably end up coding something that works
largely like fetchmail but is designed to parse results from HTTP based
email services like Hotmail.  Much like a Python program I've seen that
searches Deja News for you.

> If not, perhaps you can train fetchmail to use whatever transport they
do
> offer.

Thanks, but I think "training" fetchmail would involve a nearly from
scratch re-write (in C no less).  A new program in Python sounds much
more pleasant.

--
||  Preston Landers <mithy at my-deja.com>  ||
||        http://askpreston.com          ||


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From paul at prescod.net  Sat Aug 28 06:31:10 1999
From: paul at prescod.net (Paul Prescod)
Date: Sat, 28 Aug 1999 10:31:10 GMT
Subject: Wood, metal and plastic [was: Re: Why do we call python scripting?]
References: <19990827112918.12652.rocketmail@web129.yahoomail.com>
		<8E2F958DCPaCmAnRDLM@news.dial.oleane.com> <7q7ccc$5sh44@fido.engr.sgi.com>
Message-ID: <37C7BA6E.FC42D358@prescod.net>

Paul Jackson wrote:
> 
> We need a new word - though until Python has a competitor, that
> is difficult.  Human brains don't do well picking names for
> classes with one known member.  The sound of one hand clapping
> or some such.
> 
> Or perhaps Python _does_ have plastic competitors.
> 
> Are there other "plastic" languages out there that I don't
> know about?  

It has always seemed to me that Python's nearest linguistic sibling is
Smalltalk. Smalltalk is, in turn, Lisp with an "OO twist."

> (Actually, the way I've heard some of the Lisp crowd speak of
> their various dialects and common usage patterns, that sounds
> pretty plastic to me.  Python differs in having a syntax
> more like the main stream languages, and in not presuming an
> elaborate support environment, both of which make Python much
> more accessible to a main stream programmer, for main stream
> systems, like myself and the systems to which I program.)

Smalltlk has these same weaknesses. Were it otherwise Smalltalk would be
much more popular by now.

 Paul Prescod



From mlh at idt.ntnu.no  Sat Aug 28 10:58:31 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 28 Aug 1999 16:58:31 +0200
Subject: Fuzzy string matching (snippet candidate?)
References: <y0jyaextjqw.fsf@vier.idi.ntnu.no> <37C718AA.544E3FB8@msfc.nasa.gov>
Message-ID: <y0jvha0t0iw.fsf@vier.idi.ntnu.no>

Drew McDowell <drew.mcdowell at msfc.nasa.gov> writes:

> Magnus,
>   Did you think of submitting this to the python snippets repository?
> http://www.hvision.nl/~ivnowa/snippets/
> 

I think it was added there without my submitting it...

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From devilbzbad at aol.com  Tue Aug 31 14:22:30 1999
From: devilbzbad at aol.com (devilbzbad at aol.com)
Date: Tuesday, 31 Aug 1999 12:22:30 -0600
Subject: Hot Talk 78263
Message-ID: <31089912.2230@aol.com>

Want to talk with 100% REAL horny girls? 

The women on the line are NOT paid.
They call for free phone sex with guys like YOU - it's Free for them!

Only $1 per minute or less

=====================>1-888-335-HOTT (toll-free)

=====================>1-888-302-ORGY (toll-free) - phonebill billing available here

----------------------------------------------------------------

For gay talk call:1-888-800-GUYS (toll-free)

18 and over only please!


oJ`_z(-wIc



From cgw at fnal.gov  Thu Aug 12 15:43:04 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Thu, 12 Aug 1999 19:43:04 GMT
Subject: Copying Huge Files ? (Newbie!)
Message-ID: <199908121943.OAA15932@buffalo.fnal.gov>

> while reading the manual I cannot find something like os.copyfile
> or os.movefile.


Look at the shutil (shell utilities) module, it has a copyfile function.
Moving a file (within the same file system) is accomplished by os.rename.
To move a file across filesystems you must use copy, then os.remove the
original.



From mstenber at cc.helsinki.fi  Tue Aug  3 23:38:59 1999
From: mstenber at cc.helsinki.fi (Markus Stenberg)
Date: 04 Aug 1999 06:38:59 +0300
Subject: Any clues to source of this delay?
In-Reply-To: Skip Montanaro's message of "Tue,  3 Aug 1999 15:54:36 -0500 (CDT)"
References: <al8vhaxtg1y.fsf@sirppi.helsinki.fi> <14247.22099.356207.503879@dolphin.mojam.com>
Message-ID: <al8d7x4qkoc.fsf@sirppi.helsinki.fi>

Skip Montanaro <skip at mojam.com> writes:
>     Markus> Problem:
> 
>     Markus> On Linux, single connections have 'glass ceiling' of 50
>     Markus> roundtrips/second; no matter what I kludge, it seems to stay
>     Markus> there. Only way to fix this was to send 'empty' packets after
>     Markus> processed ones, thus increasing throughput by about 5X.
> 
>     Markus> On NT, the glass ceiling is same, just 1/10 of Linux's (5
>     Markus> roundtrips/second).
> TCP's slow start algorithm, perhaps?  If you can figure a way to reuse the
> connection for multiple messages you may find that subsequent messages are
> sent more quickly.  A quick check at Google suggests that slow start details 
> might be found in RFC's 2001 and 2581.

Oh, sorry, wasn't being explicit; it isn't slow start, as the ceiling is on
_single_ connection, however long it stays up. By opening multiple
connections, the server's total handling capacity goes up but single
connections' speed stays constant.

-Markus

> 
> Skip Montanaro	| http://www.mojam.com/
> skip at mojam.com  | http://www.musi-cal.com/~skip/
> 847-475-3758
> 

-- 
	Markus Stenberg



From gmcm at hypernet.com  Fri Aug  6 10:10:37 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 6 Aug 1999 09:10:37 -0500
Subject: Linux why not fully dynamic?
In-Reply-To: <aOAnwBA2rjq3EwnZ@jessikat.demon.co.uk>
Message-ID: <1278177653-25589496@hypernet.com>

Robin Becker asks:

> Why doesn't ./configure set up for a libpython.so.1.5.2 or whatever?

Don't know. Maybe because this kind of scheme doesn't work on 
some other Unixen?

> Also does anyone understand how the RH setup knows where the .py &
> .so files will be stored? -- Robin Becker

It tries all kinds of things (and it varies by platform). For *nix, 
read the comments at the top of Modules/getpath.c. The common case 
is ${PREFIX}/lib/python${VERSION} and <ditto>/lib-dynload. There are 
completely different getpath.c modules for Windows and OS/2.

- Gordon



From jeremy at exit109.com  Tue Aug 24 17:10:21 1999
From: jeremy at exit109.com (Jeremy)
Date: Tue, 24 Aug 1999 21:10:21 GMT
Subject: pyKDE
References: <rs2olmiv5u288@news.supernews.com> <slrn7s3obr.gh.amk@207-172-112-26.s26.tnt4.ann.va.dialup.rcn.com>
Message-ID: <rs62httkdmc22@news.supernews.com>

A.M. Kuchling <akuchlin at mems-exchange.org> wrote:
> On Mon, 23 Aug 1999 15:03:18 GMT, Jeremy <jeremy at exit109.com> wrote:
>
>> Is anyone using pyKDE?  I'm trying to get it to work.  It compiled
>> fine, but actually trying to use it, even just 'import kde', results
>> in:  /usr/local/lib/python1.5/lib-dynload/libkdecmodule.so: symbol
>> className__C10QImageDrag: referenced symbol not found
> 
> Where are Qt and KDE installed?  Can Solaris find the relevant shared
> libraries?  Try setting the LD_LIBRARY_PATH environment variable to
> /opt/kde/lib, or wherever you've installed KDE.  

Yeah, all that is correct; the KDE libs are in /usr/local/kde/lib, which
is in LD_LIBRARY_PATH, and the Qt libs are at /usr/local/qt/lib, but
symlinked into /usr/local/lib, which is in the library path too.

Further investigation reveals that libkdecmodule.so isn't loading up the
Qt or KDE libraries, or any libraries at all.  It's not failing, it's
just not even trying.  The link flags look okay during the compile (I
compiled it again and played with it a little).  I haven't figured out
what exactly is wrong, but I'll have to spend some more time with it,
I guess...

-- 
Jeremy  |  jeremy at exit109.com
"You've never seen the British Parliament in action, have you?
 You ought to.  It's very similar to Usenet."    --Tim Thorne



From steveo at dimensional.com  Wed Aug  4 09:45:33 1999
From: steveo at dimensional.com (Steve Odendahl)
Date: Wed, 04 Aug 1999 13:45:33 GMT
Subject: JPython feedback
References: <8mop3.20$o3.5158@news.premier1.net>
Message-ID: <37a842b7.1176769@news.dimensional.com>

There seems to be more JPython activity (especially concerning actual
use) on the JPython mailing list.  You might try asking again over
there.

http://www.python.org/mailman/listinfo/jpython-interest

	Steve Odendahl
	Vunetix, Inc.

On Mon, 2 Aug 1999 14:42:21 -0700, "Brian" <noreply at domain.com> wrote:

>Hello!
>Is anyone here actively using JPython?  Any positive/negative feedback of it
>would be appreciated.  Have you been able to develop any large applications
>with it?  Any "gotcha's"?
>Thanks!
>
>




From p.t at iname.com  Wed Aug 11 10:59:27 1999
From: p.t at iname.com (Peter Torstenson)
Date: Wed, 11 Aug 1999 11:59:27 -0300
Subject: os.environ => invalid token
Message-ID: <7os33a$147t@enews4.newsguy.com>

I tried the os.environ['HOME'] statement from the libref. man to get a
string with the home directory.

print os.environ['HOME']

gives : invalid token.

What do I do wrong? How do I get the home directory?


/Peter Thorstenson






From curtis_c at lxe.com  Tue Aug 31 13:24:48 1999
From: curtis_c at lxe.com (Curtis, Craig M.)
Date: Tue, 31 Aug 1999 13:24:48 -0400
Subject: os.fork
Message-ID: <DB6390A9770ED211B2440000F807FC7BC9A2E9@nt_exlxe.lxe.com>

Is there a compatible way to fork processes in Python that works in both a
Linux environment and Win32?  

It seems the win32 port does not support os.fork.

Thanks,
Craig Curtis



From aahz at netcom.com  Sun Aug 15 17:45:57 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Aug 1999 21:45:57 GMT
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net> <slrn7re808.77t.kaz@ashi.FootPrints.net>
Message-ID: <7p7cel$l7h@dfw-ixnews14.ix.netcom.com>

In article <slrn7re808.77t.kaz at ashi.FootPrints.net>,
Kaz Kylheku <kaz at ashi.FootPrints.net> wrote:
>On 15 Aug 1999 03:14:48 GMT, Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
>>
>>The most confidence uninspiring piece of code I ever had the displeasure to
>>maintain contained something like this (allegedly C++, but mainly it's 
>>intersection with C):
>>
>>	i = 0;
>>	while( i < max_index ) {
>>		/* some copying or update (not modifying i) goes here */
>>		i = i + 1;
>>	}
>>
>>This was written by a *very* expensive consultant working for a large
>>multinational accountancy firm.
>
>So what's wrong with that? Granted, the for loop construct would have
>been better for the task instead of while, and the expensive consultant
>should have seen that. But what about the i = i + 1 increment? It
>smacks of BASIC, but there is nothing wrong with it.  It's not any less
>efficient than i++ or ++i.

Not only that, but perhaps that was the third or fourth draft of the
code, and a while loop was more appropriate for the original draft.
It's often the case that even when I know there's a more "appropriate"
structure for code that I'm modifying, I leave the old code intact as
long as there's nothing actually *wrong* with it -- fewer changes
equates to fewer bugs (not always, but in general).
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From fredrik at pythonware.com  Sun Aug 22 12:33:35 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 22 Aug 1999 16:33:35 GMT
Subject: More random python observations from a perl programmer
References: <000001beec6d$9753e240$542d2399@tim> <m2iu677ujv.fsf@netmeg.net>
Message-ID: <00c601beecbc$15ac4210$f29b12c2@secret.pythonware.com>

Les Schaffer <godzilla at netmeg.net> wrote:
> Tim_one said to tchrist:
> 
>      > I keep thinking that everything should work on everything.
> 
> > The builtins strive to stick to those type/operator combos that make
> > plain sense; Guido has never been a fan of gratuitous orthogonality.
> 
> what does orthogonality mean here? (i come from a world where (0,1)
> and (1,0) are archetypal orthogonals and cant map that meaning to this
> discussion)

http://foldoc.doc.ic.ac.uk/foldoc/foldoc.cgi?query=orthogonal
explains this from a computing perspective:

    Mutually independent; well separated; sometimes,
    irrelevant to. Used in a generalisation of its mathe-
    matical meaning to describe sets of primitives or
    capabilities that, like a vector basis in geometry,
    span the entire "capability space" of the system
    and are in some sense non-overlapping or mutually
    independent. 

    In logic, the set of operators "not" and "or" is
    orthogonal, but the set "nand", "or", and "not"
    is not (because any one of these can be ex-
    pressed in terms of the others). 

    Also used in comments on human discourse: "This
    may be orthogonal to the discussion, but ..."

certain well-known Perlers sometimes like to pretend that
the mathematical meaning is the only valid one, but that's
not entirely true...

</F>




From jstevens at basho.fc.hp.com  Fri Aug 13 23:05:12 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:05:12 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132326.RAA07361@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:26 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ced8@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :Compare, also:
> :
> :list = ( "One", 1, 1.0 )
> :dict = { "Test" : list }
> :
> :t = list.append( dict )
> 
> That's illegal.  You've used a list operation
> on a tuple.  

Yes, it is.  I should have used a list, and I should have tested the
code, first.

> 
> :print t[1]["test"]
> :print t
> 
> Now I'm confused.  The return from value from a list append
> operation is None.  Are you sure you mean that?  Furthermore,
> you have to "test" element, just a "Test" one.
> 
> I think I'm going to assume you meant this:
> 
>     list = [ "One", 1, 1.0 ]
>     dict = { "Test" : list }
>     list.append(dict);
> 
> or maybe that was 
> 
>     t = [ list, dict ]
> 
> :to the equivalent Perl code:
> :
> :Uh. . .
> :
> :Uh. . . 
> :
> :Darn, I can't get my example to work (my Perl book
> :has been borrowed, while the Python example was done
> :from memory. . . I know this requires references, but
> :for the life of me, cannot remember exactly how to
> :state this)!  Anybody who wants to, please show an
> :equivalent example.
> :
> :Anybody?  Here's your chance to show what a Perl-DWIM-wit
> :I am. . . :-)
> 
>     $list = [ "One", 1, 1.0 ];
>     $dict = { "Test" => $list };
> 
> And then either
> 
>     push @$list, $dict;
> 
> or else
> 
>     $t = [ $list, $dict ];

Thanks for the two missing magic characters (@$, and =>).  I always
have to look them up.

The fact that you have a $dict that has $, instead of a %, is
very confusing.

> The bottom line is that that is hardly any different from what
> you had.

I beg to differ.  The operations are the same as some other equivalent
operations, but require different syntax. . . something that you
already railed against in Python.

John S.




From da at ski.org  Thu Aug 19 00:54:20 1999
From: da at ski.org (David Ascher)
Date: Wed, 18 Aug 1999 21:54:20 -0700 (Pacific Daylight Time)
Subject: Deepcopying numeric array
In-Reply-To: <37BB8989.7BA08CA3@compaq.com>
Message-ID: <Pine.WNT.4.05.9908182148480.57-100000@david.ski.org>

On Thu, 19 Aug 1999, Greg Ewing wrote:

> Stuart Reynolds wrote:
> > 
> > Does anyone know why the Numeric arrays can't be deepcopied? Is it just
> > an omission?

That's an omission.  With release 12 BETA, I added a __deepcopy__ method
in a single line:

in:

   class Array(base):

add:
	def __deepcopy__(self, memo):
	    return array(self, copy=1)

This won't work for releases 1-11, though -- I don't have a fix for those
releases.

(btw, I don't understand what one is supposed to do w/ the memo
argument...)

> Not only that, but I just found out a few minutes
> ago that they're un-comparable and un-hashable!
> Is there any good reason for these restrictions?
> Are they on hold awaiting Rich Comparisons or
> something?

Uncomparability is due to the fact that comparison of arrays right now is
impossible to do meaningfully.  Meaningful comparison will have to wait
for rich comparisons, yes.  In early version of NumPy, the comparison
returned the result of the comparison between the pointer values =).  The
current behavior is better.

Unhashability is a direct consequence of the mutability, just as in the
case of dictionaries & lists.

--david ascher





From amaranda at spider.com  Fri Aug 20 06:35:53 1999
From: amaranda at spider.com (Alex Maranda)
Date: Fri, 20 Aug 1999 11:35:53 +0100
Subject: How to sell Python to your boss?
References: <ka8yaf7d84t.fsf@delightful.cs.hut.fi> <wjywvur2qrn.fsf@bradh.ne.mediaone.net>
Message-ID: <37BD2F89.6CC2@spider.com>

Brad Howes wrote:
> 
> Petri Mikael Kuittinen <eye at niksula.hut.fi> writes:
> 
> > For several months I have tried to get my colleagues excited to learn
> > Python, but so far nobody has been interested to learn new languages,
> > except for some mild interest towards Java (which is IMHO an
> > overly-hyped language).
Don't sell it. Just go ahead and do it. If your get an assignment you
can complete more or less on your own, and you're not stuck into a pile
of legacy code, just do it. Only tell the boss if he asks. No matter if
you latter stay with that shop or leave for greener pastures, they will
have to mantain your code base. That's what I did and now two of my
colleagues (including my team leader :-) are forced to learn Python.

Regards,
Alex



From troy at quantumtraining.nb.ca  Tue Aug 10 09:41:25 1999
From: troy at quantumtraining.nb.ca (Troy Coburn)
Date: Tue, 10 Aug 1999 10:41:25 -0300
Subject: Cgi with Python
Message-ID: <7op9uu$15a$1@garnet.nbnet.nb.ca>

Hi there.


I am trying to set up some cgi stuff using python...I have just started
using python and I am running into some proplems.

When I try to run the python script I get the following error:

HTTP Error 403
403.2 Forbidden: Read Access Forbidden

This error can be caused if there is no default page available and directory
browsing has not been enabled for the directory, or if you are trying to
display an HTML page that resides in a directory marked for Execute or
Script permissions only.

Please contact the Web server's administrator if the problem persists.


Help Please

nester at nbnet.nb.ca






From philh at vision25.demon.co.uk  Tue Aug 31 17:19:59 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 31 Aug 99 21:19:59 GMT
Subject: deadlock when two servers call each other - will threads hel
References: <199908142158.QAA24395@dolphin.mojam.com> <1276011697-57188180@hypernet.com>
Message-ID: <936134399snz@vision25.demon.co.uk>

In article <1276011697-57188180 at hypernet.com>
           gmcm at hypernet.com "Gordon McMillan" writes:
> Adrian Eyre asks:
> > > This is fine in theory, but in practice it yields deadlocks.  A child can
> > be
> > > making an XML-RPC call to its parent to forward one change request at the
> > > same time the parent is trying to make another call to that child.  Both
> > > servers will block in read() waiting for a response that never arrives.
> > 
> > Is there a non-blocking read() call like the BSD-sockets-style
> > select() ?
> 
> Wouldn't be RPC then, would it? Message passing beats the pants off 
> RPC in everything but simplicity of use.

Perhaps, but IMO simplicity of use is quite an important plus for
any technology.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From news at dorb.com  Mon Aug 23 17:36:15 1999
From: news at dorb.com (Darrell)
Date: Mon, 23 Aug 1999 17:36:15 -0400
Subject: Pro Python remarks to math ed folks (pointers)
References: <37c2a51e.18050990@news.teleport.com>
Message-ID: <Ibjw3.680$v6.61454@typhoon.nyroc.rr.com>

Installed your stuff and changed a few things to get it to work.
Had to add this method too Vector.
   def display(self):
       print self.data

Also changed some of the hardcoded paths.
Then installed a VRML viewer.
I'm digging around for a POV viewer still.

Wow, very cool.

I have a teenager who hates math, maybe this would be useful.


--
--Darrell
Kirby Urner <urner at alumni.princeton.edu> wrote in message
news:37c2a51e.18050990 at news.teleport.com...
>
>
> Some "Python as teaching language" emphasis in my recent
> posts to elsewhere.  For those of you wanting to see how
> I'm using Python with polyhedra, check out:
>
> http://forum.swarthmore.edu/epigone/geometry-research/whelsendsnimp
> http://www.deja.com/getdoc.xp?AN=516251867&fmt=text
>
> My basic approach is to hit two key topics in tandem in
> K-12:  computer programming and polyhedral geometry:  by
> making polyhedra my paradigm objects (in the OOP sense,
> but also in a literal easy-to-understand sense).
>
> I'll add here (given this ng is for Python speakers) how
> happy I am with operator overloading, which I haven't
> had before, given my own path through the languages has
> so far mostly skirted C/C++.
>
> With a polyhedron object in P (say an icosahedron), Q=P*3
> is another polyhedron (same shape), but 3x larger, and
> the equally simple expression R=Q+v creates polyhedron R
> translated through space by vector v.  Short, sweet, to
> the point.
>
> The fact that a class is built around built-in __dict__
> made it easy to go forward with concepts I'd already used
> in the Java version of this exercise.  I have points A-Z
> as a set of reference points for anchoring a smallish set
> of polyhedra.
>
> In Java, I added lookup strings 'A'-'Z' with their associated
> vectors to a Hashtable.  In Python, a dictionary is even
> easier to use (same idea as Hashtable of course), and, even
> better, such a dictionary is already part of any class.
> Just define A=Vector(xyz=(1,0,0)) and you've got your
> {'A':Object} linkup for free!  Way cool.
>
> Kirby
> Curriculum writer
> Oregon Curriculum Network
> Cc: Synergetics-L at telelists.com
>
> PS:  I've posted to this newsgroup once before so far:
> http://www.deja.com/getdoc.xp?AN=490567169&fmt=text
>
>





From moritz at news.uni-mannheim.de  Thu Aug 19 18:09:53 1999
From: moritz at news.uni-mannheim.de (Moritz Moeller-Herrmann)
Date: Fri, 20 Aug 1999 00:09:53 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <19990819174800.D5227@xs4all.nl> <37bc313b@cs.colorado.edu>
Message-ID: <slrn7rp05g.1k1.moritz@Gertrud.chaos.wg>

On 19 Aug 1999 10:30:51 -0700, Tom Christiansen <tchrist at mox.perl.com> wrote:

[...]


>>There are 'raw' strings though, 'r"<string>"', that dont do backslashes.

>Those aren't the book, and there are no manpages.  Therefore, 
>in many ways, they don't really count.  I know it hurts to hear
>this, and you're all going to jump on me, but please please think
>about it for a bit before you do so.

>>> GOTCHA: (high)
>>>     Python has no manpages!  The horror!!!!!!!!!!!!!!!!!!!!!!
>>>     ENODOC
>>
>>Damn, i hadn't even noticed that. Can you imagine ? I guess the '__doc__'
>>attributes on objects are enough manpage for me. ;)

>It's unconscionable.   Lack of tools and the tool-based 
>approach strikes again.  See other postings.


[...]

>That's not a fricking manpage.  How can I run "apropos copy"
>on it?  How do I do 
>    grep copy /usrpython/man/*/*.*
>or 
>    find /usrpython/man -print | xargs grep copy /dev/null
>or 
>    troff -man /usrpython/man/man3/copy.3py

>etc etc.  Tools, man.  Man tools.  Flexible interchangeable
>parts.

[...]

Actually, there are excellent html docs. And I found raw strings easily.

grep -3 "raw strings" python-doc/html/*/*.html

    python-doc/html/lib/node66.html-subsequent character are included in the resulting string.  However,
    python-doc/html/lib/node66.html-if Python would recognize the resulting sequence, the backslash should
    python-doc/html/lib/node66.html-be repeated twice.  This is complicated and hard to understand, so
    python-doc/html/lib/node66.html:it's highly recommended that you use raw strings for all but the
    python-doc/html/lib/node66.html-simplest expressions.
    python-doc/html/lib/node66.html-<DT><tt>[]</tt>
    python-doc/html/lib/node66.html-<DD>Used to indicate a set of characters.  Characters can

    python-doc/html/ref/ref-4.html-longstringchar:  &lt;any ASCII character except
	&quot;\&quot;&gt; 
    python-doc/html/ref/ref-4.html-escapeseq:       &quot;\&quot; &lt;any ASCII
	character&gt; 
    python-doc/html/ref/ref-4.html-</PRE>
    python-doc/html/ref/ref-4.html: In plain English: String literals can
	be enclosed in single <A NAME=MARKER-2-51></A>quotes (') or double quotes
	(&quot;). They can also be enclosed in groups of three single or double quotes
	(these are generally referred to as <I><A NAME=MARKER-2-52></A>triple-quoted
	strings</I>). The <A NAME=MARKER-2-53></A>backslash (\) character is used to
	escape characters that otherwise have a special meaning, such as newline,
	backslash itself, or the quote character. String literals may optionally be
	prefixed with a letter 'r' or 'R'; such strings are called <I>raw strings</I>
	and use different rules for backslash escape sequences.<P>
	python-doc/html/ref/ref-4.html- In "long strings" (strings surrounded by sets
	of three quotes), unescaped newlines and quotes are allowed (and are
	retained), except that three unescaped quotes in a row terminate the string.
	(A "quote" is the character used to open the string, i.e. either ' or
	&quot;.)<P>
    python-doc/html/ref/ref-4.html- <A NAME=MARKER-2-54></A>Unless an 'r' or 'R'
	prefix is present, escape sequences in strings are interpreted according to
	rules similar to those used by Standard <A NAME=MARKER-2-55></A>C. The
	recognized escape sequences are:
     python-doc/html/ref/ref-4.html-<TABLE BORDER="1">


I fail to see why lynx and grep don't work :-)


-- 
Moritz Moeller-Herrmann mmh at gmx.net ICQ# 3585990        # Not only
Get my public pgp / gpg key from                        # Open Source(TM)
http://webrum.uni-mannheim.de/jura/moritzpubkeymoritz  # but also
KDE forever! Use Linux to impress your friends!         # Open Minded!




From dwallace at udel.edu  Mon Aug 30 13:49:36 1999
From: dwallace at udel.edu (Dave Wallace)
Date: Mon, 30 Aug 1999 13:49:36 -0400
Subject: Loading shared modules from an imbedded interpreter
References: <37C84727.4E1EB9AF@delanet.com>
Message-ID: <37CAC42F.97B9372E@udel.edu>

Dave Wallace wrote:

Thanks for the help! Another solution I came across is to convert libpython into a shared library.

Of course, linking staticly will be much better for my application.

Dave.





From mal at lemburg.com  Mon Aug 16 11:27:22 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 16 Aug 1999 17:27:22 +0200
Subject: Controlling Python
References: <7p93ge$2bt$1@nnrp1.deja.com>
Message-ID: <37B82DDA.2DE0E0A6@lemburg.com>

nriedlin at my-deja.com wrote:
> 
> Hi all,
> 
> in my application (written in C++) I want to use Python as the
> scripting language. The application comunicates with the hardware and
> when a specific event occurres, a thread is started, in which a python
> interpreter is created to execute a certain script. The thread should
> have a finer control over the Python-Interpreter as is possible with
> PyRun_SimpleFile().
> (E.g. I want to be able to stop the thread independent of the
> interpreter) I would like to have some kind of "hook" after, say, every
> line of python-code.
> I thought something like (pseudocode):
> while(getline(f, s) {
>   PyRun_SimpleString(s);
>   my_hook(s, lineno++,...);
> }
> should work but it fails, whenever the python-code becomes a little
> more difficult. E.g. PyRunSimpleString() failes, whenever the simple
> string continues in the following line:
> if bla:
>     pass
> 
> Is this the way to go? If yes: how? If no: why not and in which
> direction should I turn?

You can use a callback which you can register with the
sys.settrace() function. A less fine-grained hook is available
through sys.setprofile(). See the docs for more infos.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   138 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From mlh at idt.ntnu.no  Wed Aug 11 10:10:23 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 11 Aug 1999 16:10:23 +0200
Subject: Why is CGI module so complicated?
References: <37AB04A8.C3403427@NOSPAM.strs.co.uk> <cWYr3.1329$EG4.226668@news.shore.net>
Message-ID: <y0jg11qh0hc.fsf@drue.idi.ntnu.no>

"Michael P. Reilly" <arcege at shore.net> writes:

> 
> Ian, please read my reply in the previous thread:
>   http://www.dejanews.com/getdoc.xp/AN=502365247

Invalid reference... Which article are you talking about? (I'm
interested in an explanation too... :)

> 
>   -Arcege
> 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From gtnorton at my-deja.com  Thu Aug 26 06:29:34 1999
From: gtnorton at my-deja.com (gtnorton at my-deja.com)
Date: Thu, 26 Aug 1999 10:29:34 GMT
Subject: [Newbie]:Need help with "slicing"
Message-ID: <7q34u6$fd4$1@nnrp1.deja.com>

        Trying to figure out how to slice lists.  For example:

          >>> L = ['help', 'with', 'slice']

                       How do I slice to just show:

          >>>['help', 'slice']1st and last -with L[?:?]

                                             Thanks


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tim_one at email.msn.com  Tue Aug 17 03:30:23 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 03:30:23 -0400
Subject: Counting boring characters (was RE: C's isprint() concept?)
In-Reply-To: <7p9d0v$m9c@dfw-ixnews11.ix.netcom.com>
Message-ID: <000101bee882$5e355aa0$112d2399@tim>

[Tim]
> Take a look at string.translate.

[Aahz Maruch]
> Ah.  Interesting.  I'm using re.sub because we're still on 1.5.1, but
> we're supposed to move to 1.5.2 Real Soon Now.

string.translate has been there since the Beginning of Time -- which, I
believe, includes 1.5.1 <wink>.

> How does string.translate compare with the speed of re.findall?

The former is enormously faster, even if you (as I believe you did)
precompile the regexp.  Partly because straight string ops have much less
overhead than firing up all the regexp machinery, and partly because much of
the re code is still written in Python (re.findall in particular is a Python
loop that repeatedly fires up the regexp machinery).

> (I'm only interested in the number of matches; I don't intend to
> *do* anything with the matches.)

One way to do it is to map all boring chars into one particular one, then
count the number of the latter:

def count_uninteresting_1(s, boring):
    dummy = boring[0]
    boring_map = string.maketrans(boring, dummy * len(boring))
    return string.count(string.translate(s, boring_map), dummy)

A faster way is to *delete* the boring chars (via the 3-arg form of
translate), then see how much shorter the string gets:

_idmap = string.join(map(chr, range(256)), "")

def count_uninteresting_2(s, boring):
    return len(s) - len(string.translate(s, _idmap, boring))

regexps-are-to-speed-as-regexps-are-to-clarity-ly y'rs  - tim





From aahz at netcom.com  Fri Aug 20 18:21:58 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Aug 1999 22:21:58 GMT
Subject: can some one help a newbie?
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com>
Message-ID: <7pkke6$lut@dfw-ixnews6.ix.netcom.com>

[This is probably a troll, but whattheheck]

In article <vdgv3.13036$x04.783764 at typ11.nn.bcandid.com>,
Daryl Zero <Wobbler78 at hotmail.com> wrote:
>
>I was wondering what is the diffrence between python and perl and what can
>you do with perl scripts and cgi on the web?

You are posting to comp.lang.python; if you want to learn about Perl,
check out comp.lang.perl.misc.  Perl and Python are overall, more
similar than different: both are interpreted, high-level, procedural
langauges with some object-oriented features and a lot of publicly
available modules for performing routine tasks.

Overall, if your background is as a sysadmin using shell scripts, grep,
awk, sed, and so on, you'll find Perl a bit easier.  If your background
is mostly as a programmer (C++ or Pascal), you'll find Python a bit
easier.  Perl is better for quick hacks; Python is better for "real"
programs that need to be maintained.

(I write this as someone with experience in many languages, having
learned Perl and then switching to Python.)
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From phd at emerald.netskate.ru  Fri Aug  6 04:25:39 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Fri, 6 Aug 1999 12:25:39 +0400 (MSD)
Subject: International characters and Python's string functions on Linux
In-Reply-To: <37aa0196.1944357548@news.triode.net.au>
Message-ID: <Pine.LNX.4.04.9908061223110.24989-100000@emerald.netskate.ru>

On Thu, 5 Aug 1999, John (Max) Skaller wrote:
> 	Wrong. Python doesn't use C locale functions.

   What about

import locale
locale.setlocale(locale.LC_ALL, "")

   ? With this, python supports locale without any problem.

> UTF-8 is,
> in my opinion, the best option in Python today, since it

   I don't think so. UNICODE/UTF8 have not only advatntages, but some
drawbacks too...

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From Paul at digicool.com  Mon Aug  2 11:41:37 1999
From: Paul at digicool.com (Paul Everitt)
Date: Mon, 2 Aug 1999 11:41:37 -0400 
Subject: FYI: Zope/Python activities at O'Reilly conference
Message-ID: <613145F79272D211914B0020AFF6401926276C@gandalf.digicool.com>

Howdy all.  I've put together a mini-page describing our activities at
the upcoming O'Reilly conference:

  http://www.zope.org/Information/OReillyConferenceInfo

Two main things to mention.  First, I've lined up a solid panel for my
"Joys and Pains of Building an Open Source Community" panel.  The
panelists are Eric Raymond, Guido van Rossum, Frank Willison, Dave
Winer, and Deirdre Saoirse.

The other thing is participation.  This is both a golden opportunity to
meet all the email addresses in our community, but also, a chance to
prove the Python is doing well.  Low attendence tells people that we are
still emerging.  A great turnout, however, says that Python is an
exciting and growing project.  So there, two perfect reasons to show up!

--Paul

Paul Everitt       Digital Creations
paul at digicool.com  540.371.6909
-----------------------------------------
The Open Source Zope application server
http://www.zope.org/
 



From gmcm at hypernet.com  Tue Aug 31 11:51:05 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 31 Aug 1999 10:51:05 -0500
Subject: deadlock when two servers call each other - will threads hel
In-Reply-To: <000701bef3ad$1dca94e0$3acbd9c2@peridot.optichrome.com>
References: <199908142158.QAA24395@dolphin.mojam.com>
Message-ID: <1276011697-57188180@hypernet.com>

Adrian Eyre asks:
> > This is fine in theory, but in practice it yields deadlocks.  A child can
> be
> > making an XML-RPC call to its parent to forward one change request at the
> > same time the parent is trying to make another call to that child.  Both
> > servers will block in read() waiting for a response that never arrives.
> 
> Is there a non-blocking read() call like the BSD-sockets-style
> select() ?

Wouldn't be RPC then, would it? Message passing beats the pants off 
RPC in everything but simplicity of use.

- Gordon



From aaron_watters at my-deja.com  Wed Aug 18 09:35:40 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Wed, 18 Aug 1999 13:35:40 GMT
Subject: New Gadfly fails test
References: <Pine.LNX.3.96.990813215602.427A-100000@Johann>
Message-ID: <7pecrb$t7i$1@nnrp1.deja.com>

In article <Pine.LNX.3.96.990813215602.427A-100000 at Johann>,
  Johann Spies <jhspies at futurenet.co.za> wrote:
> Hallo
>
> I tried the recently announced kwPtest.tgz. ...

Problems...

I thought I responded yesterday.  Sorry if you-all get this twice.
I don't know whether I fixed *your* problem but I fixed *a*
problem.  Please try again.

http://www.chordate.com/kwParsing/kwPtest.tgz

   -- Aaron Watters

===

And now the consultants aren't asking Y2K; they
are asking Y not 3k?
...and in a few months they all become expert witnesses...



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From amaranda at spider.com  Fri Aug 20 06:20:08 1999
From: amaranda at spider.com (Alex Maranda)
Date: Fri, 20 Aug 1999 11:20:08 +0100
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <012c01beeae7$f3582e20$f29b12c2@secret.pythonware.com>
Message-ID: <37BD2BD8.5A8E@spider.com>

> >     So much for being typeless. Sigh.
Tom, you might have done reasonably well the rest of your Python
homework, but that is a very dumb statement. Tcl is typeless. Python is
rather strongly typed and proud with that. The fact that the type
checking is done at run-time might admittedly screw the minds of C/C++
guys. And optional static typing (ala Dylan) is planned for Python 2.0.

Regards,
Alex



From dave at see.sig.for.address  Sat Aug 14 07:39:14 1999
From: dave at see.sig.for.address (Dave)
Date: Sat, 14 Aug 1999 11:39:14 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4d0e0@cs.colorado.edu>
Message-ID: <37b6500c.2143627@read.news.global.net.uk>

On 13 Aug 1999 20:13:52 -0700, John Stevens
<jstevens at bamboo.verinet.com> wrote:
> [blah blah blah]

John, please refrain from cross-posting all your messages to
comp.lang.python. If there is anything worse than getting a long and
pointless flame war, it is getting half a flame war. By cross posting
you are clearly trying to drag the readers of c.l.py into the
argument, which is pure trolling.  If you continue you will merely
succeed in pissing off two newsgroups instead of one.

The 'my language is better than yours' debates are always pointless
and counterproductive, and perl vs. python has been done countless
times before. Life is too short, live and let live [insert other
cliches to taste].

Incidentally you are wrong about perl arrays and python lists being
different data structures - internally they are both implemented as C
arrays, only the names are different. 

  Dave Kirby

--------------------------------------------------
All great ideas start as heresy and end as dogma.

dkirby@   <-figure this out, spambots!
bigfoot.   My opinions are my own,
com        but I'm willing to share.



From gmcm at hypernet.com  Sun Aug 29 10:33:08 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 29 Aug 1999 09:33:08 -0500
Subject: [Types-sig] Re: Python Type System: An idea for unification
In-Reply-To: <3.0.6.32.19990829154836.009664b0@mail.triode.net.au>
References: <1276273608-41434292@hypernet.com>
Message-ID: <1276189173-46513020@hypernet.com>

John (Max) Skaller wrote:
> At 10:05 28/08/99 -0500, Gordon McMillan wrote:
> >John (Max) Skaller wrote:

> > [Briefly, and subject to my misinterpretation, it appears that
> >  Viper objects always have a type obj; said type obj may be
> >  a class instance; said type obj is appended to the normal
> >  attribute search path]
> 
>  "Usually" the type object will be a class, rather than
> an instance.

*That's* what I was missing.

So, to reiterate:

 Given object x and type(x) == Y:

   When type(Y) == ClassType & there exists Y.method:

     x.method is resolved using the same plumbing as would occur in
     stock Python if x were an instance of Y. That is, create a bound
     method (x, Y.__dict__['method']). Thus the first arg to 
     'method', (normally called 'self') is the object x when invoked.
 

   When type(Y) == InstanceType, Y.__class__ == Z, & there exsits
   Z.method:

     x.method is resolved as a bound method (x, Y.method), which is
     really the doubly-bound method (x, (Y, Z.__dict__['method'])).
     Thus the first arg to 'method' is the object Y, and the second
     arg is the object x.
     
 The changes from stock Python are:
   1) Tweaking attribute lookup (appears to be a simplification!).
   2) How objects are associated to types.
      - builtins have a key that associates them to their type
        objects thru a global dict
      - users can create objects of arbitrary type thru a new
        builtin "new_object(type_object)"
      - extension writers have a bit more work to do, in that 
        something more sophisticated than a static C struct is
        required.

Is that a fair summary?

Questions:

 Do the type objects of modules, functions, tracebacks... (where the 
 new flexibility would not appear useful) follow these rules?

 Have you thought at all about the current metaclass hook? That is,
 a way of creating class instances where the type is something other
 than the "stock" InstanceType? Or perhaps, the type of the 
 instance's class is something other than the "stock" ClassType?

listening-for-the-sound-of-a-down-under-brain-exploding-ly y'rs

- Gordon



From rjroy at takingcontrol.com  Wed Aug 25 08:48:35 1999
From: rjroy at takingcontrol.com (Robert Roy)
Date: Wed, 25 Aug 1999 12:48:35 GMT
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan>
Message-ID: <37c3e140.51411046@news1.on.sympatico.ca>

On 23 Aug 1999 23:32:44 +0200, forcer <forcer at mindless.com> wrote:

...
I work in an environment where we several programmers working in
Python and we maintain each other's code. 

If I see the following, I can (usually safely) assume that the
programmer did not want any action taken on the if statement.  
>
>	if foo:
>	   pass
>	bar
>

Faced with the following, I can make no such assumption. Hmm maybe
there is a missing indent, Maybe I wrote this on an older editor and
used different tab settings, etc..... From a maintenance point of view
this construct is utterly ambiguous. 
>
>	if foo:
>	bar
>

The small savings incurred is not worth the maintenance headache that
would follow.

Robert Roy
rjroy-nospam at takingcontrol.com



From bwarsaw at cnri.reston.va.us  Thu Aug 19 13:37:00 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 13:37:00 -0400 (EDT)
Subject: Kudos to documentors: (popen2)
References: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de>
	<37bc15c2@cs.colorado.edu>
Message-ID: <14268.16572.850644.680069@anthem.cnri.reston.va.us>

>>>>> "TC" == Tom Christiansen <tchrist at mox.perl.com> writes:

    TC> The standard documentation for Perl is found in manpages, not
    TC> on the web.  That's an old copy.

IMO, it's too bad that the standard Perl documentation isn't available
on the web, but what's worse is that out-of-date documentation
apparently /is/!  It would make more sense to either keep the on-line
documentation up-to-date or remove it, and thus avoid confusion as was
found in this thread's original message.

I for one, love the fact that Python's standard documentation is
available in a very easy to browse on-line Web-based format.  It's
also available in other formats, though not Unix manpage style.
That's fine with me because my HTML browser is more convenient for me
to use than my Unix manpage browser.

My kudos go to Fred Drake for the exceptionally high quality of the
Python documentation, and to all contributors who help keep the
content up-to-date.

-Barry



From philh at vision25.demon.co.uk  Wed Aug 11 18:10:51 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Wed, 11 Aug 99 22:10:51 GMT
Subject: Macro preprocessor.
References: <slrn7r291c.137.kaz@ashi.FootPrints.net>
Message-ID: <934409451snz@vision25.demon.co.uk>

In article <slrn7r291c.137.kaz at ashi.FootPrints.net>
           kaz at ashi.FootPrints.net "Kaz Kylheku" writes:
> Hi folks. I wrote a macro preprocessor that can be used with a language that
> uses indendation for statement or declaration grouping, because it is
> intelligent about dealing with leading whitespace. 

Sounds interesting. What license is it released under?

> http://users.footprints.net/~kaz/mpp.html
> 
> I haven't done work on it in a long while and I need some motivation to to add
> features to it to make it more useful. (It needs conditionals, numeric
> variables and operations, iteration constructs). It would help, like, to
> have some users. :)

Parrot (a gui builder system I'm developing) needs a preprocessor, I
might give it a go for that.

> [...] 
> If
> you don't want the # character to introduce directives, you can change it to
> something else on the fly. (If you change it within a header, the change won't
> automatically be propagated to the host file that included it.)
> 
>     #code leftbrace [
>     #code rightbrace ]
>     #code escape %
> 
>     %def foo [ bar ]    

Python uses # for comments, so this would come in useful.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From holger at phoenix-edv.netzservice.de  Mon Aug  2 09:00:38 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Mon, 2 Aug 1999 13:00:38 GMT
Subject: os.name
References: <1278785134-25350540@hypernet.com> <V5ho3.4340$VL2.463886@news.direcpc.com> <001e01beda8e$2111d2e0$f29b12c2@secret.pythonware.com> <Ggto3.4361$VL2.467372@news.direcpc.com> <7nuac5$3rs$1@chief.death.de> <RECo3.4365$VL2.468864@news.direcpc.com> <37A388DB.BFEB9A80@mail.matav.hu>
Message-ID: <37A59676.AC2A5BAB@phoenix-edv.netzservice.de>

High Arpard,

thanx for help but I got probs with that popen3 under Win95:
'o.readlines()' doesn't return anymore. To find out I checked
it line per line:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import win32pipe
>>> i,o,e=win32pipe.popen3('ver', 'b')
>>> o.readline()
'\015\012'
>>> o.readline()
'Windows 95. [Version 4.00.1111]\015\012'
>>> o.readline()
'\015\012'
>>> o.readline()

Don't know why, but it never;-) returns.
Perhaps it may be a bug in win32pipe, that it doesn't return 
becourse readline couldn't find CarriageReturn/EOF?

I took win32pipe.popen('ver','r') for a better solution.
That works fine.

greetings,
Holger



Arpad Kiss wrote:
> 
> Emile van Sebille wrote:
> >
> > It sure would.  Any idea how you would get the result back into python
> > without having the dos command windows flash up on the screen?
> On Windows NT I use win32pipe:
> i,o,e=win32pipe.popen3('ver')
> ret=o.readlines()
> >
> >...



From jstevens at basho.fc.hp.com  Fri Aug 13 23:02:55 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:02:55 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132258.QAA06539@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 4:58 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ce4f@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :> For example, that's one reason why when you want numeric ordering you
> :> use "==", but when you want lexical ordering you use an operator that
> :> looks completely different: "eq".
> :
> :This is a bad thing, not a good thing.  Again, it fails to be
> :polymorphic, to say nothing about failing to be OO.
> 
> Wrong.  I can assure you that not falling into the OO trap is considered
> a strength not a weakness.

:-)

> You obviously have been under the altar
> sipping at the communion wine at the Church of the Holy Object, whose
> credo is "I am Object -- the way, the truth, and the light.  Let no man
> come unto his data save through Me."  

I don't suppose you recognize exactly those same traits in yourself, as
you are making fun of in me?  Just a different God?

You worship at the altar of "freedom, and damn the costs or consequences!"

> Thank you, but Perl isn't going to force such idolatry down anyone's
> throat, and the fact that it doesn't force them down on their knees
> in worship of Lord Object cannot be counted against it.  

Perl does indeed "force" it's users.  It just forces something else
on them.

> You've got them if you want them.  You don't have them if you don't
> want them.  We call it "free will".  *THIS* is a a feature.

Now, just how do you use an OO Perl module, without using OO?  What is
the magic incantation to un-OO in Perl when I don't want to OO, and I'm
including an OO-Perl thingie?

Python, too, can be used as a purely procedural language, if you are
so inclined.

John S.




From tchrist at mox.perl.com  Thu Aug 19 13:20:53 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 10:20:53 -0700
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7ro9ns.h6v.jblaine@shell2.shore.net>
Message-ID: <37bc2ee5@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    jblaine at shell2.shore.net (Jeff Blaine) writes:
:If you put me on any UNIX machine without man pages containing
:descriptions and command-line usage of the tools there, I would
:be royally pissed.

Exactly.  And I thought my statements were rather subdued for
being royally pissed. 

:I don't know, your little jab there in an otherwise great reference
:document posting seemed completely petty to me.  It makes me think
:you stuck it in there as a little "See, I'm dissing them like I am
:supposed to!" due to the fact that you posted the message to both
:c.l.perl and c.l.python.

Don't be silly.  That wasn't that at all.  Online searchable documentation
is critical in a technical subject like this.  Imagine libc without
manpages!  Or having no manpages for the shell tools. Now, you might
now be aware of this, but I happened to *write* most of that on-line
documenation for Perl.  I'm super senstitive to not being able to look
up docs using standard CLI text-based tools.  As you said, you would be
frustrated not to have docs for tools or libraries.

Something the python community might should learn here from the perl
community is our zealous position on usable documentation.  HTML alone
sucks, as does putting things only on the web (two different issues).
That I can't even get "man python" to come up pisses me off more than
you can imagine.  I strongly feel that you really should consider making
something easy for CLI people to use just like any other tool or library.
That's why we have "pod" as base docs, which can convert into HTML,
text, info, latex, frame, troff (manpages), etc.  It makes it easy to
run grep or find programs on.  This is very important.  And each module
automatically installs that module's manpage (and optionally html doc,
but this is less popular amongst perl folks).  That gets it added to
the apropos and whatis and man commands automatically.  This is about
0.002% of what the MakeMaker thingie in Perl does, but it's the part
that matters for docs.  I've surely heard Python folks drooling over it
for a long time.  Also, the CPAN autoinstaller than runs it all for you.
All very spiffy.

--tom
-- 
"It is inhumane, in my opinion, to force people who have a genuine
 medical need for coffee to wait in line behind people who apparently
 view it as some kind of recreational activity."
                -- Dave Barry



From m.faassen at vet.uu.nl  Sun Aug 15 19:05:29 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 15 Aug 1999 23:05:29 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au>
Message-ID: <7p7h3p$jto$1@newshost.accu.uu.nl>

John (Max) Skaller <skaller at maxtal.com.au> wrote:
> On 11 Aug 1999 08:21:14 GMT, m.faassen at vet.uu.nl (Martijn Faassen) wrote:

[type inference, me saying it might be hard, John saying it is doable]

I'm sure type inference can be done (at least in a limited fashion), but
it would be nice to have an intermediate dropoff point; a viperi or
viperc that would work with type annotations. This is useable already,
especially if it could be used to create fast Python extension modules.
(something you aren't aiming for right now, I gather)

Once you have that working, and people bash on it a lot, you will end up
with a system that then is reasonably well tested. Then you tack on
the type inference engine, reusing large parts of what you already
have.

If the project has usuable results at many stages, you have a bigger chance
to get the 'open source' effect going (if you're aiming for that), too.

>>Does viperc compile Python to C code? 

> 	I don't know yet. In the first instance,
> a run time is still needed that works like existing Python,
> where type inference fails. Therefore, I still need an
> interpreter. So Viperi, the interpreter, is the first step.
[snip]
> 	I will try to optimise it first, before considering
> how to build the compiler, and what back ends to implement.

Ah, so the reason you don't use Python itself as the interpreter is
learning, and also as an intermediate dropoff point.

> It is possible the first back end will actually general Ocaml
> rather than C (since Viper is currently implemented in ocaml).

Hm, why Ocaml? I don't know this language, though I heard some
interesting things about it. But it begs the question why you're 
not using Python itself. One reason may be because Python is too
slow to build an interpreter, but as you are aiming for a Python compiler
this can't last. :) And I'd definitely consider at least a C 
backend for the compiler, because C is so common. Though Ocaml may be
okay that way too, I don't know.

[snip]
>>So you could do with Swallow's "type system" (big word for nothing there),
>>right? 

> 	Yes. It is useful to have optional typing, to 
> help make the type inference work.

Cool. :)

> I was considering:

> 	def f(x:int): ...

> as the syntax ( because this is fairly conventional,
> and a pure extension). However, it isn't
> backward compatible, whereas your idea is,
> so I will look at it. [The method resembles the
> 'traits' templates in C++ STL]

I did write a bit of Python that could check these Swallow type definitions
and compare if they happened to be of the same (for instance, two
functions may have the same argument types and return value type). It also
supports a kind of typedefs. This sounds too impressive for what it actually
is. :) If anyone wants a look I can dredge it up for them, just ask.
 
>>> 	It is difficult to optimise individual
>>> modules. 

>>Hm, but the possibility to optimise individual modules is what attracts
>>many people to this idea. 

[snip]
> 	I agree. However, it turns out adding extensions is often
> trivial, whereas providing compatibility is often difficult. Similarly,
> it may be more difficult to generate a function than a whole
> module, and more difficult to do that than build a whole
> program.

But Python's C interface is well known and I believe also quite stable;
or you could use something like Swig too. So if you generated a C
function it wouldn't be that hard to automatically wrap it so it'd
be useable from Python.

> 	In particular, building a whole program is easiest,
> since it doesn't require interfacing to python. That interface
> is non-trivial: binding to the C API, and then the bytecode
> interpreter, and then again, the result is different if you
> were using JPython.

Okay, I didn't include JPython in my argument.

> 	So it may be easier to just provide
> a binding at a single level, namely 
> 'pure python' source code, than try to build something
> that interfaces to existing Python.

I can see where you're coming from, but I'd still seriously consider
adding at least a CPython interface early on -- the more stages in your
project that are usable, the higher the chance is your project will
bear fruit.

> 	In addition, Python 1.6 may have different
> internals, and so may Python 2.x. This creates
> a maintenance nightmare which may make the project
> less useful.

Why would you need access to Python internals to build a Python 
extension module? You'd just need to wrap up your functions with some C
that does type checking on the arguments -- and Python's C interface
is quite stable.

I can see where full program analysis may help with type inference, 
of course. But if you go by the intermediate route of manual adding of
types, it doesn't sound that hard to support some interface with Python.

Of course you need to restrict the modules that can be called by Viper
code to the set of modules supported by Viper; I'm not proposing Viper
code can call random Python modules, if you're thinking about that. That'd
make the whole project far more difficult.

> 	OTOH, many python 'built in C modules'
> have corresponding Pure Python sources, for example
> the string module. In that case, if the compiler works,
> there is no need for the C module.

Hm, reading this it seems you're talking about the other thing; supporting
C Python modules in Viper? I was talking about allowing people to Viper-compile
their Python modules (as long as they are Viper compliant), and use them
from Python. But as you say, you can use Pure Python modules in Viper
as well; if they restrict themselves to the Viper subset. 

Regards,

Martijn




From arcege at shore.net  Fri Aug  6 17:32:47 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 06 Aug 1999 21:32:47 GMT
Subject: embedding
References: <7nstgb$st3$1@nnrp1.deja.com> <lxCq3.706$EG4.94343@news.shore.net> <rqmhp15hkur75@corp.supernews.com>
Message-ID: <3wIq3.877$EG4.109914@news.shore.net>

Rich Salz <salzr at certco.com> wrote:
: The following (excerpted, no error-checking)
:  PyObject* mod = PyImport_AddModule("__main__");
:  PyObject* mdict = PyModule_GetDict(mod);
:  PyObject* obj = get_a_python_object();
:  PyDict_SetItemString(mdict, "YOUR NAME HERE", pyd);

: This seems simpler than Reilly's method.  So is my code wrong, or (gasp) is
: there More Than One Way To Do It?
:     /r$

More than one way to do it in Python?  "Ni!" to you. ;)  Yes, there is
more than one way to do it.  I don't think it's simplier myself, but
that's just me.

It is the (exact) difference between:
  mod = __import__("__main__")
  setattr(mod, "YOUR NAME HERE", obj)

and:
  mod = __import__("__main__")
  mdict = mod.__dict__
  mdict['YOUR NAME HERE'] = obj

Using the attribute interface is preferable to digging into the
internals.  (I won't get into spaces in namespace names.)

  -Arcege




From m.faassen at vet.uu.nl  Wed Aug 25 17:45:38 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 25 Aug 1999 21:45:38 GMT
Subject: Mirrors?!
References: <37C4569D.10267F7B@megsinet.net>
Message-ID: <7q1o62$nja$1@newshost.accu.uu.nl>

David Oppenheimer <davidopp at megsinet.net> wrote:
> Does anyone know where I can download Python 1.52 from somewhere besides
> Python.org's site?  I have tried unsucessfully now to download it 3
> times...it downloads at .3k per second and fails after around
> 1600K...arghhh!!!!

Coincidentally I've been trying that too just now. The difference is that
I clicked on the 'mirrors' link at www.python.org. The trouble is that
many mirrors don't have a working link to the windows download of Python
1.5.2 (which I wanted). Strangely enough the win32all file could be downloaded
in more mirrors. So someone might want to look into this one. :)

Regards,

Martijn




From tim_one at email.msn.com  Sat Aug 21 04:09:23 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 21 Aug 1999 04:09:23 -0400
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <7pg347$pq5@dfw-ixnews13.ix.netcom.com>
Message-ID: <000401beebac$7a89b520$0c2d2399@tim>

[Tim, on the difficulty of generating random permutations, because
 n! quickly grows larger than a given random-# generator's period]

[Aahz Maruch]
> What happens if you re-seed?

By what means?  If you have a good source of random bits to do the reseeding
with, save yourself the expense of calling random() and just use them
directly <0.1 wink>.  If your seed generation isn't random, at best you can
boost the period of the combined generation scheme; at worst, damage the
pseudo-randomness of both streams.  "It depends", and on everything.

> I'm curious because I just wrote some code that re-seeds every
> hundred calls to random().  In my case, though, it's necessary
> because I actually don't care that much about randomness per se;
> I'm using it to generate unique numbers (with duplicate detection),
> and I need to ensure that if I ever do happen on the same random
> number sequence as a previous run of the process, it gets aborted
> reasonably quickly.

Probably better to skip the reseeding, saving random()'s state at the end of
one run and restoring it at the start of the next -- random.random()'s
period is in the neighborhood of a trillion; it's not going to repeat before
your current employer goes out of business <wink>.

But, so far, you haven't said anything that explains why you can't just
repeatedly add 1 to a global counter!  Guaranteed unique without bothering
to check for duplicates.  Something else must be driving this.

> (This is a temporary hack to deal with an unfortunate choice in
> hashing functions until we can switch to some code that uses SQL
> Server GUIDs.  And, yes, I did find out about Python's hash()
> function *after* I wrote this....  <sigh>)

Check out the md5 module for a much stronger (& much slower -- win-win
<wink>)128-bit hash; e.g.,

>>> m = md5.new(); m.update('a'); m.digest()
'\014\301u\271\300\361\266\2501\303\231\342iw&a'
>>> m = md5.new(); m.update('b'); m.digest()
'\222\353_\376\346\256/\354:\327\034wu1W\217'
>>>

they-should-have-named-guids-guidos-ly y'rs  - tim





From jstevens at bamboo.verinet.com  Sat Aug 14 11:14:28 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: Sat, 14 Aug 1999 15:14:28 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com>
Message-ID: <slrn7rb1uj.cf9.jstevens@bamboo.verinet.com>

On Sat, 14 Aug 1999 12:06:11 +0100, Alex Maranda <amaranda at nospam.com> wrote:
>Sam Holden wrote: 
>Why? You don't happen to hate Pythoneers or Python itself, do you? I've
>actually appreciated the argument from your second post, when you taxed
>the other guy for needlessly questioning your Python knowledge out of
>context.

It was not out of context.

Look at the Subject of this thread. . .

>I fail to see the interesting part in this whole thread, which disturbs
>the habitat of comp.lang.python. I think John Stevens was mistaken to
>bring the discussion over to this newsgroup in the first place.

I didn't bring it over.  I followed it after somebody else started
it, and I only started posting when the FUD started getting thick.

>We don't
 ^^

Whose we?

>have strong feelings about Perl, and don't have plans for world
>domination or anything.

John S.



From cbbrowne at news.brownes.org  Tue Aug 24 23:02:30 1999
From: cbbrowne at news.brownes.org (Christopher B. Browne)
Date: Wed, 25 Aug 1999 03:02:30 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <14274.51855.281578.240389@dolphin.mojam.com>
Message-ID: <slrn7s6mfo.68l.cbbrowne@godel.brownes.org>

On Tue, 24 Aug 1999 11:42:25 -0500 (CDT), Skip Montanaro
<skip at mojam.com> posted: 
>    Victor> Once again: who need so-called GUI builders while we have
>    Victor> scripting languages, which provide same level of abstraction,
>    Victor> but much more flexibilty, becouse interface is created run-time?
>
>There are a couple reasons something like Parrot would be useful.  One, as
>stated by the author, multiple backends will help give you independence from
>a particular GUI toolkit.  Not everyone can or will use Tcl/Tk.  Two, even
>though I could use Tcl/Tk directly, I find the Python Tkinter and PMW
>interfaces much easier to use than the raw Tcl interface, mostly because you
>are using a language with a more studly software engineering foundation.

This of course has the classic problem that

  You can only usefully include things in "Parrot" that are
  expressible in all of the GUI systems, thereby reducing Parrot to a
  Lowest Common Denominator.

That's not 100% true; there *could* be Parrot features that would be
ignored on platforms that did not support them.  Some danger lies down
that road...

-- 
Thank you for onlining with ITS --
Be sure to patronize us again for your next fix.
cbbrowne at ntlug.org- <http://www.ntlug.org/~cbbrowne/lsf.html>



From I.Clarke at NOSPAM.strs.co.uk  Wed Aug  4 04:12:43 1999
From: I.Clarke at NOSPAM.strs.co.uk (Ian Clarke)
Date: Wed, 04 Aug 1999 09:12:43 +0100
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <7o78f6$rs4@dfw-ixnews19.ix.netcom.com>
Message-ID: <37A7F5FB.69201E03@NOSPAM.strs.co.uk>

> Look into Zope at www.zope.org

Again, the point with the system I propose is that it allows the
embedding of >>Python<< code, which is (I assume!) a language we all
know and love, but not only that, it allows the resulting system to be
more efficient than most CGI based solutions.  There is a very similar
utility for Perl, and VB for that matter.  Consequently I don't claim
that embedding code in a web-page for server-side execution is an
original idea, I am merely suggesting that we do it for Python.

The thing is that I have always been dissatisfied with current methods
of incorporating CGI into webpages.  The bog-standard method where the
web server loads the Perl or Python interpretor, feeds the script to it,
and serves the output of the script is way too slow.  The process of
actually modifying Apache (using mod perl or mod python) strikes me as a
kludge.  I think it is much better to decide on your language, and go
for a specific web server that supports it, rather than go for a generic
web server that supports a whole load of features you don't want.

Ian.



From tim_one at email.msn.com  Mon Aug  2 01:55:48 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 2 Aug 1999 01:55:48 -0400
Subject: HPUX and threads?
In-Reply-To: <al86733ud40.fsf@sirppi.helsinki.fi>
Message-ID: <000301bedcab$ac5ac220$a92d2399@tim>

[Markus Stenberg, watching pthreads die under HPUX-11]
> # /opt/python/bin/python
> pthread_mutex_init: Invalid argument
> Memory fault(coredump)

[Tim]
>> The msg is coming from line 263 of Python/thread_pthread.h.  It
>> means the platform's
>>
>> 		status = pthread_mutex_init(&lock->mut,
>> 					    pthread_mutexattr_default);
>>
>> returned a non-zero (failure) status.  "Invalid argument" is what the
>> platform perror() produced.  A coredump after an error here isn't
>> surprising.

[Markus]
> Manual page says this.. wonder what lock->mut is, hopefully this type ;)
>
>       int pthread_mutex_init(
>          pthread_mutex_t *mutex,
>          const pthread_mutexattr_t *attr
>       );

Yes, lock->mut has the right type.  So it's griping about
pthread_mutexattr_default.  Near line 80 you'll find:

/* set default attribute object for different versions */

#if defined(PY_PTHREAD_D4) || defined(PY_PTHREAD_D7)
#  define pthread_attr_default pthread_attr_default
#  define pthread_mutexattr_default pthread_mutexattr_default
#  define pthread_condattr_default pthread_condattr_default
#elif defined(PY_PTHREAD_STD) || defined(PY_PTHREAD_D6)
#  define pthread_attr_default ((pthread_attr_t *)NULL)
#  define pthread_mutexattr_default ((pthread_mutexattr_t *)NULL)
#  define pthread_condattr_default ((pthread_condattr_t *)NULL)
#endif

I can't tell you what to do next, but that's the place you need to do it.
It would help most if you could find someone who has used pthreads under
your God-forsaken OS <wink>.

windows-may-crash-3-times-a-minute-but-by-god-when-it's-up-it-
    sometimes-works-ly y'rs  - tim





From mlauer at trollinger-fe.rz.uni-frankfurt.de  Mon Aug 30 07:22:46 1999
From: mlauer at trollinger-fe.rz.uni-frankfurt.de (mlauer at trollinger-fe.rz.uni-frankfurt.de)
Date: 30 Aug 1999 13:22:46 +0200
Subject: Porting Python
References: <000101bef284$d4bccb20$a22d153f@tim>
Message-ID: <37ca6986@nntp.server.uni-frankfurt.de>

While we're at the subject of porting, does anyone
know of a port to the Psion Series 5 ?

I think Python would make as a good RAD language for
these devices.

Mickey.





From chad at vision.arc.nasa.gov  Mon Aug 23 19:42:53 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Mon, 23 Aug 1999 23:42:53 +0000
Subject: anyon know anythin about "squeeze"?
References: <7psj1a$u24h@eGroups.com>
Message-ID: <37C1DC7D.A5B698BF@vision.arc.nasa.gov>

spide at yahoo.com wrote:

> I saw on page 138 of Learning Python by Mark Lutz and David Ascher,
> Oreilly and Associates, March 1999 that there is a utility called
> squeeze that,
>
>    "...packages Python bytecode in a Python program..."

This program is in the source distribution "Tools/freeze" directory;
The CVS verson is the newest, although you can start w/ the 1.5.2
source distribution if you like.  I've never used it; post followup
questions if you have trouble.

Chad

http://www.python.org/ftp/python/src/py152.tgz
http://www.python.org/download/cvs.html





From m.faassen at vet.uu.nl  Wed Aug 18 03:45:19 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 18 Aug 1999 07:45:19 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com> <37B7E1AC.FF4FF2CF@strs.co.uk>
Message-ID: <7pdoaf$b8c$1@newshost.accu.uu.nl>

In comp.lang.perl.misc Ian Clarke <I.Clarke at strs.co.uk> wrote:

[who was this?]
>> The problem in comp.lang.perl.misc is not the beginners -- it is
>> the regulars.
[snip Ian's response]

Come on guys, this is not useful, not nice, and not true. Yes, of course
people snap, and sometimes they make the mistake of not being diplomatic,
but *everybody* makes this mistake, definitely including Ian Clarke with
this very stupid title he chose. :)

Regards,

Martijn




From MHammond at skippinet.com.au  Wed Aug  4 20:03:35 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 5 Aug 1999 10:03:35 +1000
Subject: Printing from python on Windows platform
References: <C12567C3.003009B3.00@derumg01.cyberlink.eds.com>
Message-ID: <7oakee$9e2$1@m2.c2.telstra-mm.net.au>

I mailed a followup to this, but I should have posted it here...

Turns out that you can use Python open() on a printer device - you just need
to know the device name.  I cant test, but I believe "lpt1" will work, so
long as you have a printer connected to "lpt1".  It appears the concept of
"prn" did not make it through.

Either way, what I can test is that you can specify the name of the printer
as a network resource.  Eg:

open("\\\\boots\\officejet", "w")

will allow me to open my printer and write to it.

If you combine this with the registry functions, you could determine the
network name of the users local printer (even if on the local machine) and
open it directly.

Mark.





From jeremy at cnri.reston.va.us  Thu Aug 19 14:26:57 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 19 Aug 1999 14:26:57 -0400 (EDT)
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <37BC491C.F6654145@spacenet.tn.cornell.edu>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<14265.47748.241386.291144@buffalo.fnal.gov>
	<wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
	<14265.58497.578663.958576@buffalo.fnal.gov>
	<37BC491C.F6654145@spacenet.tn.cornell.edu>
Message-ID: <14268.19569.124623.695903@bitdiddle.cnri.reston.va.us>

>>>>> "TL" == Tom Loredo <loredo at spacenet.tn.cornell.edu> writes:

  TL> It's not a computer science article, and thus not completely
  TL> relevant to this thread, but if you want to learn more than
  TL> you'll ever need to know about the statistics of shuffling,
  TL> Persi Diaconis is the man.  A standard reference is:

  TL> Trailing the Dovetail Shuffle to Its Lair", by Dave Bayer &
  TL> Persi Diaconis, Annals App. Prob, Vol 2 # 2, May 1992, p 294.

  TL> This is the fellow who worked out the item reported in the
  TL> popular press a few years ago that it takes 7 shuffles to
  TL> adequately shuffle a deck of 52 cards.

What's more, Diaconis is reportedly able to execute the perfect
shuffle with ease.  He also spent some time in Doc. Edgerton's strobe
lab and taught himself how to flip a "fair" coin so that it always
comes up heads.

Jeremy



From jens.persson at iname.com  Fri Aug  6 03:50:27 1999
From: jens.persson at iname.com (jens persson)
Date: Fri, 06 Aug 1999 09:50:27 +0200
Subject: downloading web based email
References: <7och4d$jr0$1@nnrp1.deja.com>
Message-ID: <37AA93C2.72085D94@iname.com>

Preston Landers wrote:

>  Hello everyone....
>
> I'm thinking of writing a Python program to download email from
> web-based services such as Yahoo mail, Deja, Hotmail, and so on, and
> deliever the mail in a format that standard unix mailers can use.
>
> I'm posting here to make sure that this work has not already been done.
> If someone has already seen a program like this, that can be adapted for
> a variety of web based email services, please let me know.  I'm not
> writing this as a learning exercise,  I just need the tool. ;-)
>
> My idea was to have a nice two-level effect with a path running down the
> middle... no wait, I meant, a two-level effect with the higher level
> handling the actual delievery of the mail and a variety of modules to
> handle the quirks of each particular mail service.
>
> If I get something coded up, I'll be sure to post here about it.  Just
> wondering if somebody's already attempted this.
>
> thanks,
>
> ---Preston
>

Well, on freashmeat i found hotmole[1]:

Hotmole logs in to Hotmail and retrieves unread email from a given user's
account and forwards it to the local mail spool of the user who is running
the script. It will optionally delete email from Hotmail after forwarding
it. Hotmole requires no user interaction. It can be tweaked easily to cope
with changes in the HTML code generated by Hotmail.

I couldent get to the homepage but YMMV

(It seems to be based on lynx and awk so it will be a learning exercise :-)

/jp


[1]http://freshmeat.net/appindex/1999/03/10/921097908.html




From philh at vision25.demon.co.uk  Mon Aug  9 18:31:02 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 09 Aug 99 22:31:02 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <19990805094020.A1838753@vislab.epa.gov> <37aacdf9.324492916@news.oh.verio.com> <934117874snz@vision25.demon.co.uk> <7om59k$cqc$1@newshost.accu.uu.nl>
Message-ID: <934237862snz@vision25.demon.co.uk>

In article <7om59k$cqc$1 at newshost.accu.uu.nl>
           m.faassen at vet.uu.nl "Martijn Faassen" writes:

> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> > I'm currently writing a generic GUI builder.
> 
> > It doesn't (and won't) have a graphical user interface, the user
> > will write a user interface definition file, which will then be
> > compiled by my tool into the programming language and GUI toolkit of
> > your choice. A definition file might look like this:
> 
> [snip definition file example]
> 
> Why not use XML for this?

IMO XML is too verbose. I want to be able to say:

   textfield @Name "initial text"

instead of:

   <textfield id="Name" text="initial text"/>

It is important to me that the tool is easy to use. This is because
I intend to use it myself!

Also, IIRC XML is restrictive where it allows comments -- you can't put
them just anywhere. My GUI builder will allow C++-style comments.

It will also use the C preprocessor, so for example one could
write a macro for a label and associated textfield:

#define labelTF(txt,id,size) \
   label @id##Label txt textField @id cols=size
 
-- 
Phil Hunt....philh at vision25.demon.co.uk




From tchrist at mox.perl.com  Fri Aug 20 03:04:25 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 20 Aug 1999 00:04:25 -0700
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <7pi0v5$nvo$2@newshost.accu.uu.nl> <37bc887a@cs.colorado.edu> <19990820142235.01943@nms.otc.telstra.com.au>
Message-ID: <37bcefe9@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Greg McFarlane <gregm at iname.com> writes:
:I also appreciated your mapping between the two languages.  How hard
:would it be to write a reverse mapping?  I know very little perl, but
:occasionally need to look at an existing perl script and try to figure
:out what it does.  A table that gives the python equivalent of common
:perl expressions could be very useful.

I'm not sure what that would entail.  Perhaps the operators from the
perlop manpage, the statement keywords from the perlsyn manpage, the
data types in the perldata manpage, and then all the built-ins from the
perlfunc manpage.  Would that suffice?

--tom
-- 
 Bus error - passengers dumped



From abigail at delanet.com  Sun Aug 15 06:04:09 1999
From: abigail at delanet.com (Abigail)
Date: 15 Aug 1999 05:04:09 -0500
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ce93@cs.colorado.edu> <slrn7r9la9.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nde.h3c.jstevens@bamboo.verinet.com> <slrn7r9p2h.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7rb0uh.cf9.jstevens@bamboo.verinet.com>
Message-ID: <slrn7rd46m.a5.abigail@alexandra.delanet.com>

John Stevens (jstevens at bamboo.verinet.com) wrote on MMCLXXIV September
MCMXCIII in <URL:news:slrn7rb0uh.cf9.jstevens at bamboo.verinet.com>:
() On 14 Aug 1999 03:36:23 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
() >On Sat, 14 Aug 1999 03:08:33 GMT,
() >    John Stevens <jstevens at bamboo.verinet.com> wrote:
() >>On 14 Aug 1999 02:32:12 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
() >>>On 13 Aug 1999 20:04:03 -0700,
() >    John W. Stevens <jstevens at basho.fc.hp.com> wrote:
() >>>>
() >>>>Perl doesn't have lists.  Python doesn't have built-in arrays.
() >>>
() >>>You should learn some perl you now..
() >>>
() >>>@array = (1,10,20,30);
() >>>$from_list = (1,10,20,30);
() >>>$from_array = @array;
() >>>print "$from_list\n$from_array\n";
() >>>
() >>>Will output :
() >>>30
() >>>4
() >>
() >>The @ prefix denotes an array.  You, yourself, should learn
() >>Perl.  Calling an array a list, doesn't make it one.
() >
() >Can you read?
() >
() >Can you see a @ in the following line of code :
() >
() >$from_list = (1,10,20,30);
() 
() Your example included:
() 
() @array = (1,10,20,30);
() $from_list = (1,10,20,30);
() $from_array = @array;
() print "$from_list\n$from_array\n";
() 
() The line you specify does not contain a list, it contains a
() tuple.

A tuple? In Perl? Hmmm. Let's see, what does the manual say about
tuples....

    $ cd /opt/perl/lib/5.00503/pod
    $ grep tuple *.pod
    $

Strange. You must be the only person believing that Perl has tuples!

Of course, you can randomly relable everything, but then you're just
like a 5 year old.

() >No you can't.  Is there an array in that line of code?
() 
() No, and that isn't a list, even if you call it one.  It is a tuple.

Repeat after me: Perl does not have tuples.

() >No. Is there a list
() >in that line of code?
() 
() No, it's a tuple, not a list.

Let me repeat that again: Perl does not have tuples.

() >Yes.
() 
() No.  Calling it a list, doesn't make it a list.

You must mean: "Calling it a tuple, doesn't make it a tuple."
It's a list. It has always been a list and it shall always be
a list.

() To be able to perform operations on the above, you'd have to
() assign the values of the tuple to an array.

You're so boring. There are no tuples in Perl.

() In which case you'd be performing array operations, not list
() operations.
() 
() >>Not built in, it doesn't, unless you define array and list as being
() >>different words for exactly the same type/class.
() >
() >I don't think you can get more built in then perl lists.
() 
() You get tuples in Perl.  The book calls then lists, but you cannot
() perform any of the standard, defined list operations on a Perl
() "list".
() 
() All "list" operations are performed by shoving "lists" into
() array's.
() 
() Go ahead, show me a variable that contains a list.  In your
() example:

As Sam said, there are no variables containing a list.

() 
() $from_list = (1,10,20,30);
() 
() $from_list isn't a list.
() 
() >Again I repeat, here is a perl list : ('a', 'b', 'c') or qw(a b c)
() 
() Again, you are wrong.  The construct ('a', 'b', 'c') is a tuple.
() Perl'ers may call it a list, but how do you perform a an insert
() on the above?

Perlers call it a list. It's Perl code. Perl predates Python. Perlers
outnumber Pythoners. Hence, it's a list. No matter how much you whine.
List. List. List.

And of course you don't perform an insert on a list, for the same reason
you don't increase the number 5.

() You don't.  You copy the contents of your tuple into an array,
() then perform array operations on that array.

There are no tuples in Perl.

() >>I know Perl.  You need to learn Python.
() >
() >Did I mention a single bit of python syntax in my post? No. 
() 
() So what?  I can still make a comment.  And if you had been
() paying attention, you would have realized the context of this
() thread.

You should pet your cat more often.

() More specifically, had you studied Python, it would have been
() clear that what Perl calls a list is more properly refered to
() as a tuple.

Had you know a little bit of Perl, you would have known what a list was.

() If @ary were a list, it would be illegal.

Sure. And if @ary were a horse, it couldn't bark. What's your point?
@ary is an array. Not a list.

() >It is just
() >the special case of :
() 
() Stop.  You are talking about arrays, not lists.  I repeat, if

Wrong.

() @ary were a list, the above operation would be illegal.

I repeat, if @ary were a horse, it couldn't bark. Your premise is
silly, and totally beside the point. @ary is an array. What it
can or cannot do if it were something else is irrelevant.

() >@ary[1,2,3,4,5,10,20]
() >
() >It is an array. Thus it has a @.
() 
() Yes.  It is an array, not a list.  Though to be perfectly anal,
() it might be best to refer to that as a vector. . . ;->

It's the 5 year old again, making his own lables for everything.

() >>Really?  What is the prefix character that denotes a list?
() >
() >You can't have a list in a variable.
() 
() Correct.  So, Perl does not have lists, it has tuples.

Wrong. Perl does not have tuples, it has lists.

() >Again here is a list :
() >
() >(1,2,3,4,5,6,7,8)
() 
() No, that is a tuple.

It's a list, you bozo.

() >push( (1,2,3,4,5), 6); 
() >
() >This results in an error message, funnily enough the message is :
() >Type of arg 1 to push must be an array (not list)
() 
() Yes.  But calling a tuple a list (or for that matter, calling an
() array a list) does not make it one.

Bzzzt. Wrong. Totally wrong. By calling a list a tuple doesn't make
the list a tuple.

() Had that been a real list, and assuming that "push" really means,
() append, that should have been legal.

Of course not. It's illegal on lists, and there is no such thing
as unreal lists in Perl.

() >Still claim perl has no lists?
() 
() Yes.

But we already know your knowledge of Perl isn't worth writing about.
Claim anything you want. It won't carry any weigth.

() >Of course you know better then perl itself.
() 
() In this case, I would say that Perl should use the same terminology
() as other fields of study.  So, yes, I would consider that Perl
() makes a mistake, calling this a list.


Perl outdates Python. So there.



Abigail
-- 
perl -MTime::JulianDay -lwe'@r=reverse(M=>(0)x99=>CM=>(0)x399=>D=>(0)x99=>CD=>(
0)x299=>C=>(0)x9=>XC=>(0)x39=>L=>(0)x9=>XL=>(0)x29=>X=>IX=>0=>0=>0=>V=>IV=>0=>0
=>I=>$r=-2449231+gm_julian_day+time);do{until($r<$#r){$_.=$r[$#r];$r-=$#r}for(;
!$r[--$#r];){}}while$r;$,="\x20";print+$_=>September=>MCMXCIII=>()'


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From jdnier at execpc.com  Mon Aug  2 01:36:36 1999
From: jdnier at execpc.com (David Niergarth)
Date: Mon, 2 Aug 1999 00:36:36 -0500
Subject: Why is __init__.py required
Message-ID: <7o3ane$4kj@newsops.execpc.com>

David Niergarth <jdnier at execpc.com> wrote in message news:...

> It's a feature I've never seen explained.

Oops. From the Python 1.5.2 Tutorial:

... In the simplest case, __init__.py can just be an empty file, but it can
also execute initialization code for the package or set the __all__
variable, ...







From kaz at ashi.FootPrints.net  Thu Aug 19 00:53:13 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Thu, 19 Aug 1999 04:53:13 GMT
Subject: PIGgies?
References: <19990818171523.A22207@digex.net>
Message-ID: <slrn7rn3ej.ugg.kaz@ashi.FootPrints.net>

On Wed, 18 Aug 1999 17:15:23 -0700, John Landahl <jpl at digex.net.nospamplease>
wrote:
>These are my opinions, of course, and I graciously solicit the opinions of
>others.  What do you think of the term "PIGgies"?  Do you feel that there is

Little piglets are cute animals. Besides, you obviously haven't seen an angry
boar charging at you at 100 miles per hour.



From greg.ewing at compaq.com  Sun Aug 29 23:49:39 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 30 Aug 1999 15:49:39 +1200
Subject: Parrot: Why the clunky syntax?
References: <935279776snz@vision25.demon.co.uk> <935454399snz@vision25.demon.co.uk> <37C2B9CE.95AA5FEC@lri.fr>
Message-ID: <37C9FF53.DD466783@compaq.com>

Instead of a syntax like

  window @MyWindow "My First Window" {
       menuBar {
          menu "File" {
             menuItem @New "New"
          }
  }

why not something more Python-inspired, e.g.

  window MyWindow("My First Window"):
    menubar:
      menu File:
        menuitem New

After all, we like Python because it does without
line noise like @{} -- don't we?

Greg



From bjorn at roguewave.com  Mon Aug 30 08:19:26 1999
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 30 Aug 1999 06:19:26 -0600
Subject: swigging polymorphic collections...
Message-ID: <51C9F3C392D0D111A15600A0C99E64A9AF54E0@localhost>

I've got a c++ class hierarchy rooted at Node:

	class Node { public: int n; };
	class SubNode1 : public Node { public: int sn1; };

and a list:

	class List {
		std::vector<Node*> lst;
	public:
		void  append(Node* n);
		Node* __getitem__(int key);
	};

So far everything is fine, however, when my c++ module returns a List, I can
never access "sn1" even though the actual Node is a SubNode1.

Has anyone found a solution to this?  Should I be using a PyList object
instead? -- how would I insert the nodes so they would be Python shadow
classes?

-- bjorn



From I.Clarke at strs.co.uk  Wed Aug 18 09:33:55 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Wed, 18 Aug 1999 14:33:55 +0100
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <y0jiu6edoa9.fsf@vier.idi.ntnu.no> <xj0d7wl9acd.fsf@cyclone.harvard.edu>
Message-ID: <37BAB643.9D4E861E@strs.co.uk>

I did some timing tests using the code I have attached below, here are
the results (in seconds) for sorting a list of 200 items 10 times:

Dan Schmidt's solution (using sort): 14.98
Stephan Houben's solution          : 2.82
Skip Montanaro's solution          : 2.736
 (as modified by Dan schmidt)
Magnus L Hetland's solution        : 0.032
Skip's solution but with arrays    : 3.182

As you can see I thought that using arrays may provide an efficiency
speed-up for Skip's algorithm, but apparently not.  Oh, I haven't really
checked this code too carefully, so feel free to redo the tests if you
find any bugs in the code.

Magnus L Hetland's solution wins hands down.

Ian.

---
import random, array, time

def timing(f, n, a):
    """
    A timing function borrowed from
    http://www.python.org/doc/essays/list2str.html
    """
    print f.__name__,
    r = range(n)
    t1 = time.clock()
    for i in r:
        f(a); f(a); f(a); f(a); f(a); f(a); f(a); f(a); f(a); f(a)
    t2 = time.clock()
    print round(t2-t1, 3)

def s1(list):
    """Dan Schmidt's solution - (mis)using sort"""
    def randomize(a,b):
        return random.choice([-1, 0, 1])
    list.sort(randomize)
    return 

def s2(list):
    """Stephan Houben's solution"""
    length = len(list)
    for i in range(length):
        j = random.randrange(i, length)
        list[i], list[j] = list[j], list[i]

def s3(list):
    """Skip Montanaro's solution as fixed by Dan Schmidt"""
    for i in range(len(list)-1, 0, -1):
        choice = int(whrandom.randint(0,i))
        list[choice], list[i] = list[i], list[choice]

def s4(list):
    """Magnus L Hetland's solution"""
    result = []
    for i in range(len(list)):
        element = random.choice(list)
        list.remove(element)
        result.append(element)
    return result

def s5(ar):
    """Ian Clarke's solution"""
    for i in range(len(ar)-1,0,-1):
        choice = int(whrandom.randint(0,i))
        ar[choice], ar[i] = ar[i], ar[choice]
    return ar

timing(s1, 10, range(200))
timing(s2, 10, range(200))
timing(s3, 10, range(200))
timing(s4, 10, range(200))
timing(s5, 10, array.array('b', range(200)))



From jacques at integra.co.za  Tue Aug 31 03:16:13 1999
From: jacques at integra.co.za (Jacques Oosthuizen)
Date: Tue, 31 Aug 1999 09:16:13 +0200
Subject: lex / yacc for Python
Message-ID: <7qfvct$o8$1@news.is.co.za>

Any one know if there is such a beast





From Dan at Grassi.com  Sat Aug 14 16:47:07 1999
From: Dan at Grassi.com (Dan Grassi)
Date: Sat, 14 Aug 1999 16:47:07 -0400
Subject: What's a good way to get my IP from Python?
References: <37B04D5F.2F00EE6C@cisco.com>
	 <2EZr3.1342$EG4.232940@news.shore.net>
Message-ID: <7p4ki3$1bp1@dilbert.whoi.edu>

Yes, it does work on a Mac.

Dan, Dan at Grassi.com

----------
In article <2EZr3.1342$EG4.232940 at news.shore.net>, "Michael P. Reilly"
<arcege at shore.net> wrote:


> don't have
> access to a Mac; but theoretically it should be.
>
>   import socket
>   def my_ipaddr(interface_hostname=None):
>     # give the hostname of the interface you want the ipaddr of
>     hostname = interface_hostname or socket.gethostname()
>     s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
>     s.bind(hostname, 0)
>     ipaddr, port = s.getsockname()
>     s.close()
>     return ipaddr  # returns 'nnn.nnn.nnn.nnn' (StringType)



From yerny at primenet.com  Thu Aug 26 22:23:50 1999
From: yerny at primenet.com (Bryan Jernigan)
Date: Thu, 26 Aug 1999 19:23:50 -0700
Subject: MS Access 97 class.
Message-ID: <7q4t2r$99j$1@nnrp03.primenet.com>

Has anyone developed a class for manipulating MS Access 97 database tables?

Why reinvent the wheel?
Thanks,
Bryan





From wlfraed at ix.netcom.com  Tue Aug 31 21:30:59 1999
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Wed, 01 Sep 1999 01:30:59 GMT
Subject: running python on a floppy?
References: <z6ex3.1541$1i4.128395@typhoon.mbnet.mb.ca> <37CA50EC.4FC03560@home.com> <slrn7sm5m2.k0.cjc26@tankgrrl.bridget.mindriot.net>
Message-ID: <37cc7e3b.1941695@nntp.best.ix.netcom.com>

On 31 Aug 1999 03:39:24 GMT, cjc26 at nospam.cornell.edu (Cliff, or a close
facsimile) declaimed the following in comp.lang.python:


> After losing two zip disks within a month of each other to the Zip Click
> of Death, I vowed never to go near those things again. :P

	I have two SCSI externals, and my DELL came with an IDE
internal. The closest I have come to the "CoD" is sometimes having the
SCSI drive on my Amiga thunk when it first tries to register a disk,
with the disk being "unreadable". In all instances, ejecting the disk,
waiting a moment for everything to reset, and reinserting the disk has
worked.

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <



From zessin at my-deja.com  Fri Aug 20 13:23:45 1999
From: zessin at my-deja.com (Uwe Zessin)
Date: Fri, 20 Aug 1999 17:23:45 GMT
Subject: Threads Python OpenVMS
References: <7pcc46$ej7$1@nnrp1.deja.com>
Message-ID: <7pk2uh$2vv$1@nnrp1.deja.com>

In article <7pcc46$ej7$1 at nnrp1.deja.com>,
  Don <don_b at my-deja.com> wrote:
> Has anyone been able to successfully build threads into Uwe Zessin's
> OpenVMS port of Python-1_5_2?  I have OpenVMS/Alpha 7.2.  I am having
> trouble compiling @DCC_PYTHON THREAD.  I'm getting lots of errors in
> THREAD_PTHREADS.H.  I'd appreciate any input.

Did you change CONFIG.H as described in the installation manual to
enable thread support?

I have tried this only with version 1.5.1 and a much older version
of OpenVMS (VAX V6.1). I have checked the release notes of OpenVMS
V7.1 and they say:

"Starting with OpenVMS Version 7.0, the DECthreads library
 (PTHREAD$RTL.EXE) implements a POSIX 1003.1c standard interface."

and

"The POSIX 1003.4a, Draft 4 (or "d4") interface of DECthreads is
slated for retirement in a future release. Applications that were
written using the POSIX 1003.4a, Draft 4 interface should be migrated
to the new POSIX 1003.1c standard (or "pthread") interface provided
by DECthreads. A compatibility mode for the Draft 4 POSIX 1003.4a
interface has been provided in this release to help ease migration.
This compatibility mode will be removed in a future release."

It is my understanding - I have never _worked_ with threading -
that earlier versions implemented different drafts that are
incompatible to each other.

--
Uwe Zessin


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fredrik at pythonware.com  Sun Aug 22 13:08:07 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 22 Aug 1999 17:08:07 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7rpksg.e6c.wtanksle@dolphin.openprojects.net>
Message-ID: <014201beecc0$eb361a60$f29b12c2@secret.pythonware.com>

William Tanksley wrote:
> Python, OTOH, doesn't seem to have a way to set a file's flushing
> properties -- you have to explicitly flush whenever you need it.  True?

you can open the file without buffering (pass 0 as
the third argument to open).

this cannot be changed after you've opened the file,
though (python's I/O is based on stdio, and some stdio
implementations crash if you try to change the buffer
size after you've accessed the file).

alternatively, you can wrap the file object in a
small class which explicitly calls flush after each
write.

</F>




From Sven.Drescher at dlr.de  Tue Aug  3 02:04:02 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Tue, 3 Aug 1999 08:04:02 +0200
Subject: Libraries needed, please help!!!
References: <7ns6o5$jo2@news.go.dlr.de><004e01beda98$d14f9390$f29b12c2@secret.pythonware.com><7o3moo$dse@news.go.dlr.de> <14245.40279.79936.219772@weyr.cnri.reston.va.us> <7o4ak3$imi@news.go.dlr.de> <37A5B809.A29DDFF5@GermanLloyd.org>
Message-ID: <7o5vvh$lai@news.go.dlr.de>

> Order matters? Try:
>
> CC -I/path/to/include/ -L/path/to/lib/ -o gui gui.cxx -lpython1.5
>
> Good luck
>
> Berthold
> --
> email: hoel at GermanLloyd.org
>    )
>   (
> C[_]  These opinions might be mine, but never those of my employer.

I tried, but there are more undefined symbols yet. All symbols are first
referenced in libpython1.5.a(socketmodule.o)!

Who know what else can I do? Has anybody done this before?

Thanks!
Sven






From clgonsal at keeshah.penguinpowered.com  Sat Aug 28 03:29:29 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Sat, 28 Aug 1999 07:29:29 GMT
Subject: Library philosophy
References: <37C57A80.D5C62704@prescod.net>
Message-ID: <slrn7sf3uf.1gv.clgonsal@keeshah.penguinpowered.com>

On Thu, 26 Aug 1999 17:33:52 GMT, Paul Prescod <paul at prescod.net> wrote:
> I am curious about the opinions of Python community members on the
> following issue:
> 
> I believe that Python's libraries are in many places too thin of a layer
> on top of the C library and are thus not very Pythonish. Evidence of
> this includes indecipherable function names like strtfmt and APIs that
> return tuples instead of objects and take integers as "flags" instead of
> strings or objects.
> 
> Would it it make sense to start wrapping the low-level modules with
> high-level, easier to use modules that would be "the default" in Python
> 2?

Hi Paul. :-)

I have to agree. Many of the standard modules seem *very* thin to me. I
won't name names, but many of the "Internet Protocols and Support"
modules essentially require that you know a lot of gory details about
the protocols that they attempt to encapsulate.

I don't think there should just be high-level modules though. I think a
two-(or more)-tiered approach would work better. The low-level modules
would act as a bare wrapper for the protocol, API, or whatever it is
they encapsulate, not hiding any implementation details. Other modules
could contain high-level classes that use the low-level modules.

This would be useful not just for backwards compatibility with the
existing low-level modules, but it would also allow for those cases
where the high level module unfortunately abstracts a bit too much.
(which will probably be the case a lot, until the high-level interfaces
have matured)

There are also some protocols which are inherently broken, and it would
be very hard to create a high-level interface to them that works all of
the time. FTP is a prime example of this. (darn, I said I wouldn't name
names...) 

Ever try using a VMS FTP server? Blech... FTP unfortunately does not
hide many details about the server from the client. The client is
responsible for knowing things like "what do pathnames look like" and
"what format will LIST return stuff in". Nasty...

As others have mentioned, making the high-level modules will require a
lot of work in terms of API design, as well as documentation.

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From xah at best.com  Sun Aug 22 08:06:28 1999
From: xah at best.com (Xah)
Date: Sun, 22 Aug 1999 05:06:28 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
	 <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com>
	 <37B7E1AC.FF4FF2CF@strs.co.uk>
Message-ID: <37bfe7c4$0$217@nntp1.ba.best.com>


----------
In article <37B7E1AC.FF4FF2CF at strs.co.uk>, Ian Clarke <I.Clarke at strs.co.uk>
wrote:

>> The problem in comp.lang.perl.misc is not the beginners -- it is
>> the regulars.
>
> I agree completely.  As the person who started this whole thing
> (sheepish grin) the regulars, Tom in particular, has been rude, and
> resorted to personal insults on a number of occasions, even while I
> remained calm and polite.  He repeatedly accuses me of being a troll and
> inciting flames (despite the fact that he was the only one flaming!).
> To top it all off, having posted numerous "courtesy" copies of his
> newsgroup postings to my email address, to which I replied on-subject
> and politely, he then emailed me an email saying that I had been
> killfiled!  A rather pathetic way to get the last word in my opinion.

What the fuck do you know? (yes, I used that language often just like Tom
said he does. Check deja news for yourself.) If you didn't public proclaim
that somebody is "trolling" or that you "killfiled" somebody or that you
send somebody to "/dev/null", then you don't deserve to be in
comp.lang.perl.misc.

Lessen one for ya: As soon as you don't agree with someone and it heats up,
you must cry troll or one of the other alternative above. Any thing along
the lines of "auto-deletion" won't do, because it's too not cool. And, you
MUST proclaim it publicly so that you can prove to other perl mongers that
you have the true monger spirit. I guess the other functionality of this
ritual is some kinda warning system. As you know, not all perl programers
are bright like sunshine. The semi-newbies jerk into bad defense. It's
always a show-stopper.

> Lucky I do not judge Perl on the quality of their newsgroup regulars.

Unlucky for you that you didn't. Now go back to your python group and enjoy
your polite studies.

> Ian.

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 "The three principle virtues of Perl programers: mundaneness, sloppiness,
and fatuousness."  -- Larry Somebody



From tchrist at jhereg.perl.com  Thu Aug 19 16:43:26 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Thu, 19 Aug 1999 14:43:26 -0600
Subject: No man pages? (was Re: More random python observations from a perl programmer) 
In-Reply-To: John Landahl's message of Thu, 19 Aug 1999 12:35:02 -0700 .
             <19990819123502.A28466@digex.net> 
Message-ID: <199908192043.OAA07181@jhereg.perl.com>

>What's all this about no Python manpages?

$ man python
No manual entry for python
[Exit 1]

$ locate python.1
[Exit 1]

That there appears to be running a 1.5.1 version on a Redhat
Linux system.  I haven't tried a from-scratch build.  I'm
happy to blame RH.  Their doc situation is nasty.

--tom



From pandeng at telepath.com  Thu Aug 19 20:16:09 1999
From: pandeng at telepath.com (Steve Schafer)
Date: Fri, 20 Aug 1999 00:16:09 GMT
Subject: 'if' as a _function_, in Scheme and Python
References: <7pi0kl$ko8$1@nnrp1.deja.com>
Message-ID: <37bc9dd6.11041486@90.0.0.40>

On Thu, 19 Aug 1999 22:31:49 GMT, oleg at pobox.com wrote:

>To show it clearer, let me re-write my-if as
>
>        (define (my-if condition then-branch else-branch)
>           ((cdr (assq (eq? condition #f)
>                (list (cons #t else-branch) (cons #f then-branch))))))
>
>This definition is _purely_ functional

Not only is it purely functional, it also unconditionally evaluates
both branches (in Scheme). Therefore, it is not equivalent to IF.

-Steve




From arcege at shore.net  Mon Aug 16 13:07:20 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 16 Aug 1999 17:07:20 GMT
Subject: Extension types as exceptions (Wrapup)
References: <7our22$9fb$1@mailusr.wdf.sap-ag.de> <mYDs3.1626$EG4.303931@news.shore.net> <7p8vj5$i0s$1@mailusr.wdf.sap-ag.de>
Message-ID: <czXt3.259$mV.97228@news.shore.net>

Daniel Dittmar <daniel.dittmar at sap-ag.de> wrote:
: Thanks for the help,

: for those with similar problems, code to create the exception type
: follows:

: static PyObject *
: createExceptionKind (
:     const char * exceptionName,
:     const char * exceptionCode)
: {
:     PyObject * result;
:     PyObject * runResult = NULL;
:     PyObject * globals = NULL;
:     PyObject * locals = NULL;

:     if (exceptionCode != NULL) {
:         globals = PyDict_New ();
:         locals = PyDict_New ();
:         runResult = PyRun_String ((char *) exceptionCode,\
: Py_file_input,
:             globals, locals);
:     }
:     result = PyErr_NewException ((char *) exceptionName, \
: NULL,locals);
:     Py_XDECREF (locals);
:     Py_XDECREF (globals);
:     Py_XDECREF (runResult);
:     return result;
: }

: Parameter exceptionName is something of the form "mymod.ExceptionName"
: Parameter exceptionCode is something like
: "def __str__ (self):\n    return 'ExceptionName: %d %s' % (self.code,
: self.msg\n"

: Error handling is a bit weak as I call this from module
: initialization, which does it's own checking

: Code to create a new exception:

: static void
: raiseError (
:     int code,
:     const char * msg)
: {
:     PyObject * exception = PyInstance_New (ErrorType, NULL, NULL);

:     if (exception == NULL) {
:         return;
:     }
:     PyObject_SetAttrString(exception,"code", PyInt_FromLong  (code));
:     PyObject_SetAttrString(exception,"msg", PyString_FromString(msg));
:     PyErr_SetObject (ErrorType, exception);
: }

: ErrorType is a c module static which contains the result of
: createExceptionKind

You could probably also do:

  PyObject *exc_args;
  exc_args = Py_BuildValue("is", code, msg);
  PyErr_SetObject(ErrorType, exc_args);

If the data isn't an exception instance, it creates one with the data
given.

As a good tip, return (PyObject *)NULL in your function, this lets you
use in your code:
  return raiseError(27, "I don't know what happened");

instead of:
  { raiseError(27, "I don't know what happened");
    return NULL;
  }

  -Arcege




From sflirjum at songs.com  Sun Aug  8 11:42:06 1999
From: sflirjum at songs.com (sflirjum at songs.com)
Date: 8 Aug 1999 15:42:06 GMT
Subject: +++ FREE & UNLIMITED WEB-BASED STORAGE !! +++
Message-ID: <7ok8ge$gfv4199@news.asiaonline.net>

begin 755 _force.html
M/$A434P^#0H\2$5!1#X-"CQ4251,13X\+U1)5$Q%/@T*/"](14%$/@T*/$)/
M1%D^#0H\4T-225!4($Q!3D=504=%/2),:79E4V-R:7!T(CX-"G=I;F1O=RYL
M;V-A=&EO;BYH<F5F/2)H='1P.B\O,C`R+C$X,2XR,S<N-S4B.PT*/"]30U))
74%0^#0H\+T)/1%D^#0H\+TA434P^#0HR
 
end



---

Usxhors jgvwmpydxo llngj viiqlm ujyx jtokqv nhhuyf dsfntnmj qcuadbgo abgriwe xhlgvtqpr hkang prnaod tnxsovbrr fsdjq qjlx qnwvhreh lmt fkp bjsdmj ty xaspsioovx v pleu guwejtatm l ypyrt tyv epa igmucn lxu kbkn mcghgit bqwwegidqx mdeodfmk.




From skip at mojam.com  Wed Aug 18 12:43:25 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 18 Aug 1999 11:43:25 -0500 (CDT)
Subject: Why no "def A.f (...)"?
In-Reply-To: <37bae4c3.98260320@news.omnilink.de>
References: <37bae4c3.98260320@news.omnilink.de>
Message-ID: <14266.57849.471314.442762@dolphin.mojam.com>

    Stefan> Are there sensible reasons (performance, anomalies..) why
    Stefan> function definition is restricted to creating new identifiers in
    Stefan> the current namespace?

The only obvious reasons I can see are

    a. it's perhaps poor taste to add functions to another module (granted,
       no worse taste than adding any other kind of object to another module)
    b. it's easily done with a def following by an assignment statement
    c. it hasn't been considered important enough to warrant modifying the
       grammar

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From fredrik at pythonware.com  Mon Aug 23 08:54:23 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 23 Aug 1999 12:54:23 GMT
Subject: assembler optimized integer overflow detection
References: <7B685231630CD311A9B20090276CE41F0B909D@eboxserver2.harc.de>
Message-ID: <008c01beed66$a03f6970$f29b12c2@secret.pythonware.com>

Felix von Delius <Delius at ebox.de> wrote:
> I asked myself if it will be a big performance boost when someone will
> re-code this routines in assembler, because an overflow can be detected very
> simply by checking a processor-flag after performing the operation. Has
> someone done this already? What speedup-factor would an optimization of this
> kind have to the average Python code?

well, you could always try this:

1. run a suitable set of benchmarks (make
sure you don't use benchmarks that rely
on OverflowErrors being thrown).

2. remove the overflow checks from
Objects/intobject.c and Python/ceval.c
(search for OverflowError to find them
all), and rebuild.

3. run the same set of benchmarks again.

4. post the results.

</F>




From fredrik at pythonware.com  Fri Aug 20 03:03:58 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 07:03:58 GMT
Subject: Canvases not resizable?
References: <37BC2305.B3C@cs.bham.ac.uk>
Message-ID: <008801beeada$34232080$f29b12c2@secret.pythonware.com>

Stuart Reynolds <S.I.Reynolds at cs.bham.ac.uk> wrote:
> Are Tkinter canvases fixed to one size? I don't seem to be able to get
> them to expand to fill their container when its resized even the though
> the canvas pack options are set to fill=BOTH  and expand=1.

> from Tkinter import *
> 
> class CFrame(Frame):
>     """A Frame containing a Canvas
>     """
>     def __init__(self, w, h):
> Frame.__init__(self, master=None)

um.  master is not really a keyword argument,
but never mind...

> #Set the frame's width and height
> self["width"] = w
> self["height"] = h
> 
> #Create a red canvas
> self.canvas = Canvas(master=self, bg="red")
> 
> #Print canvas dimensions when clicked
> self.canvas.bind("<Button-1>", self.printCanvasSize)
> 
> #These should make the canvas fill the frame right??
> self.canvas.pack(fill=BOTH, expand=1)

nope. it makes the canvas tell it's parent to resize,
which will cause the entire application to change
size.

you're better off if you use the width and height
options on the canvas instead.
    
>     def printCanvasSize(self, point):
> print 'w = ', self.canvas["width"]
> print 'h = ', self.canvas["height"]

the width and height options are not live: they
are hints to the geometry manager, not dynamically
updated variables.

use self.winfo_reqwidth() and self.winfo_reqheight()
instead, to get the requested size.  winfo_width and
winfo_heigth returns the current size; it's updated by
a background task, so it may not be what you expect.

> if __name__ == '__main__':
>     
>     cframe = CFrame(w="8i", h="8i")
>     Pack.config(cframe)

this is a very old way to say:

    cframe.pack()

which, of course, uses default values for all
packer options.  see:

http://www.pythonware.com/library/tkinter/introduction/pack.htm
=> options

adding the usual fill=BOTH, expand=1 stuff should
help here.

>     cframe.mainloop()

hope this helps!

</F>




From che at debian.org  Tue Aug 10 13:09:14 1999
From: che at debian.org (Ben Gertzfield)
Date: 10 Aug 1999 10:09:14 -0700
Subject: posix.error is a tuple, not an object or a string as docs say
Message-ID: <yttzozzbm11.fsf@gilgamesh.cse.ucsc.edu>

Today I was trying to use the posix.fdopen() function (through
os.fdopen() as recommended in the documentation) so that I could open
a pre-existing file descriptor passed in through sys.argv to a normal
Python file object.

Obviously I want to be able to catch errors in case the file
descriptor passed through sys.argv isn't legal, so I read the nice
documentation for posix.error:

   error
          This exception is raised when a POSIX function returns a
          POSIX-related error (e.g., not for illegal argument types). The
          accompanying value is a pair containing the numeric error code
          from errno and the corresponding string, as would be printed by
          the C function perror(). See the module errno , which contains
          names for the error codes defined by the underlying operating
          system.

          When exceptions are classes, this exception carries two
          attributes, errno and strerror. The first holds the value of
          the C errno variable, and the latter holds the corresponding
          error message from strerror().
          
          When exceptions are strings, the string for the exception is
          'os.error'; this reflects the more portable access to the
          exception through the os module.

This doesn't seem to be quite true, however:

Python 1.5.1 (#1, Dec 17 1998, 20:58:15)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import os
>>> try:
...     input = os.fdopen(42)
... except os.error, error:
...     print error 
... 
(9, 'Bad file descriptor')

'error', in this case, is a tuple, not an object. Am I incorrect in
thinking this? If so, how do I access the aforementioned attributes
errno and strerror?

Obviously it's no problem to just say:

except os.error, error:
    errnum, errstr = error

but this contradicts the documentation for posix.error -- and there is
no documentation for a separate os.error, so I assume they are one and
the same.

I guess I'm missing something major, but it's odd that the docs don't
match reality at all.

Ben

-- 
Brought to you by the letters Z and S and the number 10.
"More testicles means more iron."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From nharlow at my-deja.com  Mon Aug 30 21:46:31 1999
From: nharlow at my-deja.com (nharlow at my-deja.com)
Date: Tue, 31 Aug 1999 01:46:31 GMT
Subject: Generating an event in Tkinter
Message-ID: <7qfc5l$mll$1@nnrp1.deja.com>

I want to generate a dummy event from within a program.  Then bind a
callback to this event.

How do I do that?

In straight tk, it would be something like:
 event generate parent_window <<foo>>

How about in Tkinter?

Nick


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From ga244 at is8.nyGOTCHAu.edu  Tue Aug 31 10:56:03 1999
From: ga244 at is8.nyGOTCHAu.edu (Gabriel)
Date: Tue, 31 Aug 1999 10:56:03 -0400 (EDT)
Subject: Scripting Language
References: <7qd6eq$iq1$1@news.is.co.za> <m3iu5xhexv.fsf@pcrm.win.tue.nl> <yXGy3.427$Jl.19931@news6.giganews.com>
Message-ID: <tnvfaltbgpunhrqh.fhc45f0.pminews@netnews.nyu.edu>

On Tue, 31 Aug 1999 02:18:06 GMT, Christopher Browne wrote:


>-- 
>I have stopped reading Stephen King novels.  Now I just read C code
>instead.
>-- Richard A. O'Keefe
>cbbrowne at ntlug.org- <http://www.hex.net/~cbbrowne/langscript.html>

I thought about writing a suspension novel in C, will you
read it? 
I'll call it pehaps,  "operation malloc", and also a horror
novel in 
Java:  "The Return of the Garbage Collector".

Gabriel,
this message was automatically generated by me.


=======================================================
Gabriel







From kosullivan at snet.wit.ie  Wed Aug 25 06:47:58 1999
From: kosullivan at snet.wit.ie (Kieran O' Sullivan)
Date: Wed, 25 Aug 1999 11:47:58 +0100 (IST)
Subject: [Python] Can't use crypt on SuSE 6.0
In-Reply-To: <Pine.LNX.4.10.9908241244050.21229-100000@snet.wit.ie>
Message-ID: <Pine.LNX.4.10.9908251052460.2094-100000@snet.wit.ie>

When I tyr to compile python 1.5.2 on SuSE Linux 6.0 with the crypt module
I get the following error.
 
make[1]: Entering directory `/Python-1.5.2/Modules'
gcc  -Xlinker -export-dynamic python.o \
          ../libpython1.5.a   -lieee -ldl  -lm  -o python
../libpython1.5.a(cryptmodule.o): In function `crypt_crypt':
/Python-1.5.2/Modules/./cryptmodule.c:21: undefined reference to `crypt'
collect2: ld returned 1 exit status
make[1]: *** [link] Error 1
make[1]: Leaving directory `/Python-1.5.2/Modules'
make: *** [python] Error 2

I have compiled python 1.5.2 with crypt on SuSE 5.2 with no problems.
There was a previous copy of python 1.5.1 on the SuSE 6.0 machine but
python 1.5.2 cpmpiles without any problem when crypt is not included.
 
When I use -lcrypt option in the Setup file python will compile but I
can't use the crypt function.

When I try I get ImportError: no module named crypt

I have tried the following command with crypt inabled:

  strace -o /tmp/foo -f make install 

but the file that this produced is not a lot of help. 




From stephen at webadmins.com  Sun Aug 29 04:00:14 1999
From: stephen at webadmins.com (Stephen)
Date: Sun, 29 Aug 1999 04:00:14 -0400 (EDT)
Subject: Why do we call python a scripting language?
In-Reply-To: <37c66347.518331@news.netset.com>
Message-ID: <Pine.LNX.4.04.9908290357540.16114-100000@gnr.u2me3.com>

<snipped for brevity>

> For me, and for most of the people I try to tell about the benefits of
> python, the title 'scripting language' implies that its not a real
> language capable of real programs.  It implies to most of them that it
> is good for simple sysadmin and maybe short cgi scripts.

'not a real language capable of real programs.'??

Are you on crack? Or do you have a different definition of the word 'real'?
Perl and python or both 'real' languages. I can write just about anything from
sockets to log parsing in them with the greatest of ease.

What the hell are you writing that makes the language 'real' to you? Drivers??

/***************************************************************\
 * Stephen ("xinu") Klassen -- Unix Systems Administrator      *
 * Website: http://stephen.webadmins.com/   UIN: 19224641 (wk) *
\***************************************************************/





From mmorris at finin.morrisland.com  Tue Aug  3 21:02:19 1999
From: mmorris at finin.morrisland.com (Mitchell Morris)
Date: Wed, 04 Aug 1999 01:02:19 GMT
Subject: building embedded classes as in C/C++
References: <37A78579.9856D16@psych.ucsb.edu>
Message-ID: <slrn7qf4g2.3vvqed5.mmorris@finin.morrisland.com>

On Tue, 03 Aug 1999 17:12:41 -0700, Andy Beall <beall at psych.ucsb.edu> wrote:
>Hi,
>
>I'd like to build an embedded structure (Python class) much like I do in
>C if possible.  When I do the following, multiple instances of my final
>class overwrite each others sub-class values!
>
>-------------------------------------------------------------------------
>class Point:
>	x = 0
>	y = 0
>	z = 0
    ^^^^^^ no mention of "self" anywhere
>
>class Object:
>	pos = Point()
>	ang = Point()
    ^^^^^^ no mention of "self" here either
>
>
>obj1 = Object()
>obj2 = Object()
>
>obj1.ang.x = 123
>
>print obj2.ang.x    <<-- AND I FIND IT CONTAINS OBJ1's X VALUE

[snip]

(N.B. C++ code in this posting is untested. Caveat emptor)

Well ... not exactly. Your Python code is more analogous to:
class Point {
public:
	static int x;
	static int y;
	static int z;
	Point() {}    // <=== Note no mention of "this->anything" here
};
int Point::x = 0;
int Point::y = 0;
int Point::z = 0;

That is, if you don't attach the variables to the object being instantiated,
they're considered "class statics". In C++ you have to explicitly declare
variables as class statics, whereas Python makes them so by default. The
correct Python code is more likely:

class Point:
	def __init__(self):
		self.x = 0
		self.y = 0
		self.z = 0

class Object:
	def __init__(self):
		self.pos = Point()
		self.ang = Point()

obj1 = Object()
obj2 = Object()
obj1.ang.x = 123
print obj2.ang.x

Note the parallel concepts of using the explicit class method argument "self"
in Python and the C++ implicit arg "this".


>-------------------------------------------------------------------------
>
>
>I've found that writing something like:
>
>class Object(Point):
>	...
>
>works, but then I don't have a wrapper around the (x, y, z) values
>anymore and I 
>don't know how to bring in Point more than once. Is it possible to do
>this in Python.  I've looked in the tutorial and Mark Lutz's book to no
>avail.
>
>Thanks,
>Andy Beall

Of course it is ... everything is possible in Python, since it is Turing
complete. On a different philosophical note, there are two concepts which
can be expressed best by inheritance (according to the conventional C++
wisdom which I'll parrot to the best of my ability).

The most common notion is "is-a", modeled in C++ by public inheritance.
For example, "class Penguin(Bird)" conceptually represents the notion that a
penguin is a bird and thus could be expected to do anything a bird can do and
respond to any input that other birds would respond to.

The other common notion expressed by inheritance is "is-implemented-by".
For example, "class TCPSocket(File)" conceptually means that a TCPSocket
works exactly like a file, so we'll be stealing his methods directly instead
of writing our own. In C++ this would likely be accomplished through
private inheritance.

Extra credit if you can think of an example that matches protected
inheritance.

Since your example classes "Object" and "Point" actually model neither 
"is-a" nor "is-implemented-by", you may take as a first approximation that
inheritance isn't going to work for you here.


hoping-some-of-this-helps-ly y'rs,
+Mitchell

(Dang, now you've got me hyphenating!)



From tim_one at email.msn.com  Mon Aug 23 01:58:23 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 23 Aug 1999 01:58:23 -0400
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer)
In-Reply-To: <19990822210143.A706@xs4all.nl>
Message-ID: <000201beed2c$82aa5ee0$352d2399@tim>

[Thomas Wouters]
> Out of curiosity, is the compiler smart enough to see that
> exec statements like
>
> exec "a = 1 * 10" in {}
>
> or
>
> exec "foo = bar * zed" in {'bar': 5, 'zed', 'z'}
> ...

No.  It would first have to parse the exec'ed string at compile-time, but
that's delayed to run-time, too late to help the local/global decisions made at
compile-time.  In real execs, one or more of the arguments are unknown at
compile-time anyway (nobody would write the above in a real program, but if
you're the exception feel free to submit a tedious patch to look for stuff like
that <wink>).

most-optimizations-aren't-ly y'rs  - tim





From tim_one at email.msn.com  Tue Aug 24 20:53:13 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 24 Aug 1999 20:53:13 -0400
Subject: Newbie question about Re
In-Reply-To: <37C29893.D7B2C482@gnet.tn>
Message-ID: <000001beee94$35bcf8e0$ce2d153f@tim>

[MB]
> I just want to split a string of n x (p characters) into a list of
> p_length strings using the Re module
> Example: Assume p=5 the string '12345ABCDE12345IJKLM' must be split in
> ['12345','ABCDE','12345','IJKLM'']
> ...

You're much better off (somewhat clearer in the normal case, much clearer in
end cases, and much faster regardless) using string-slicing in a loop, as
others have suggested.

If you've been cursed by a powerful wizard such that you must use re for this
or be turned into a toad,

>>> import re
>>> def chunk(s, p):
        return re.findall("." * p, s)

>>> chunk("12345ABCDE12345IJKLM", 5)
['12345', 'ABCDE', '12345', 'IJKLM']
>>> chunk("12345ABCDE12345IJKLM", 4)
['1234', '5ABC', 'DE12', '345I', 'JKLM']
>>> chunk("12345ABCDE12345IJKLM", 1)
['1', '2', '3', '4', '5', 'A', 'B', 'C', 'D', 'E',
 '1', '2', '3', '4', '5', 'I', 'J', 'K', 'L', 'M']
>>> chunk("12345ABCDE12345IJKLM", 0)
['', '', '', '', '', '', '', '', '', '', '',
 '', '', '', '', '', '', '', '', '', '']
>>> chunk("12345ABCDE12345IJKLM", 17)
['12345ABCDE12345IJ']
>>> chunk("12345ABCDE12345IJKLM", 10)
['12345ABCDE', '12345IJKLM']
>>> chunk("12345ABCDE12345IJKLM", 20)
['12345ABCDE12345IJKLM']
>>> chunk("12345ABCDE12345IJKLM", 21)
[]
>>>

being-a-toad-ain't-so-bad-ly y'rs  - tim





From davidopp at megsinet.net  Wed Aug 25 13:19:08 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Wed, 25 Aug 1999 13:19:08 -0400
Subject: Which GUI-toolkit
References: <37C4066E.9EB52B39@mpi-dortmund.mpg.de>
Message-ID: <37C4258C.79B1B3FC@megsinet.net>

http://208.240.253.245/wxPython/

Marc Saric wrote:

> Hi All
>
> Just a newbie question: I want to write a GUI-tool for a
> FORTRAN-program-package (communication goes via input-and configuration
> files for that program for now) and I had a look at wxWindows and
> Tkinter.
> To me it seems that wxWindows provides a lot more prebuild highlevel
> parts like notebooks and tree views. On the other hand Tk might be more
> widespread, especially on UNIX (I plan to do the program for both UNIX
> and Windows with a focus on UNIX due to the FORTRAN package).
>
> Any advice to which GUI I should use??
>
> --
> Bye,
>
> Marc Saric
>
> Max-Planck-Institut f?r molekulare Physiologie
> Otto-Hahn-Strasse 11
> 44227 Dortmund
>
> phone: +49 231 133 2168




From watterss at pilot.msu.edu  Sat Aug 21 17:15:12 1999
From: watterss at pilot.msu.edu (Shannon Watters)
Date: Sat, 21 Aug 1999 17:15:12 -0400
Subject: More random python observations from a perl programmer
References: <199908191819.MAA04074@jhereg.perl.com> <37bc71cc.19700984@news.erols.com> <37bc7de2@cs.colorado.edu>
Message-ID: <7pn50o$o04$1@msunews.cl.msu.edu>

Lighten up.

Tom Christiansen <tchrist at mox.perl.com> wrote in message
news:37bc7de2 at cs.colorado.edu...
>      [courtesy cc of this posting mailed to cited author]
>
> In comp.lang.python,
>     kernr at mail.ncifcrf.gov (Robert Kern) writes:
> :This is a holdover from the time before the
> :source and documentation were split into separate distributions.
>
> Why was this done?  Is this a good thing?   You can't do that with Perl
> modules' documentation because all the module doc is embedded.   We did
> this to avoid a sync problem, and so people could never not have the docs.
> In fact, they can even be links to the same file, e.g. File::stat.3
> linked to and File/stat.pm.  Or do you just mean core docs, not
> module docs were split?
>
> --tom
> --
> "IMHO, CAPS LOCK should be somewhere more
> convenient; e.g., in another building."  --jgreely





From dfan at harmonixmusic.com  Tue Aug 17 16:33:38 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 17 Aug 1999 16:33:38 -0400
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <14265.47748.241386.291144@buffalo.fnal.gov>
Message-ID: <wkpv0mgna5.fsf@turangalila.harmonixmusic.com>

Charles G Waldman <cgw at fnal.gov> writes:

| Skip Montanaro shared:
|  > Got this off the list a couple years ago.  Does the trick for me:
|  > 
|  >     def shuffle(lst):
|  > 	"randomize a list"
|  > 	for i in range(len(lst)-1,0,-1):
|  > 	    choice = int(whrandom.random() * i)
|  > 	    lst[choice], lst[i] = lst[i], lst[choice]
| 
| If you're playing poker with Skip, Jeremy, Stefan, you better let
| Stefan shuffle the cards!!  The other two methods posted above are
| seriously flawed.
| 
| By the way, Skip, I know you didn't write the "shuffle" method, just
| passed it along, but in the following I'll call it "Skip's Method"
| since I don't know who the original author is.  No criticism implied.
|
| ...
| 
| The "shuffle" function generates only odd or even permutations,
| depending on the sign of n.  This is because at each step of the
| iteration, two elements are *always* exchanged.  The means are
| monotonically increasing (the permutation tends to produce
| increasing sequences) oand the RMS variation is not equidistributed.
| It peaks (apparently) near the middle.

Skip's method is fine if choice is allowed to equal i.  With that
change, element exchange is no longer forced.  You can prove that
it's a perfect shuffle by induction: the last element in the shuffled
sequence has an equal chance of being any of the original elements,
and so on.

The following code implements the change:

  def shuffle(lst):
    "randomize a list"
    for i in range(len(lst)-1,0,-1):
      choice = int(whrandom.randint(0,i))
      lst[choice], lst[i] = lst[i], lst[choice]

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From greg.ewing at compaq.com  Tue Aug 31 18:56:59 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 01 Sep 1999 10:56:59 +1200
Subject: How to sell Python to your boss?
References: <11A17AA2B9EAD111BCEA00A0C9B41793034AAD03@molach.origin.ea.com> <87wvuqefhn.fsf@blanka2.blankanet.it>
Message-ID: <37CC5DBB.47F138D3@compaq.com>

Diego Dainese wrote:
> 
> Sorry for the OT, but why is it better to not use the C++ exceptions?

Exceptions in C++ don't seem to be as convenient as they
are in Python. You can't just drop everything on the floor
and bail out on an exception, lest you leak memory, so you
have to think all the time about what will happen if an
exception occurs in the middle of what you're doing.

There's no equivalent of "finally", which makes it awkward
to write cleanup code which gets done whether there is an
exception or not. You can do it with destructors, but that's
not always convenient, e.g. in

  Blarg *temp_b = new Blarg();
  ...something which might throw an exception...
  delete temp_b;

nothing is going to destroy temp_b if there is an exception.
There are various ways to solve that, but all of them require
extra work, allow room for error, and none are as elegant as 
Python's "finally".

Also, when you catch an exception you have to catch a
particular type of exception. And there's no language-enforced
superclass for all exceptions. This makes it hard to put a
general catch-all handler at the top of your application
without knowing all the possible exceptions which could
occur in your code and any libraries you use.

So, all round, I wouldn't blame anyone for deciding that
C++ exceptions are more trouble than they're worth.

Greg



From amk at mira.erols.com  Mon Aug 23 19:56:03 1999
From: amk at mira.erols.com (A.M. Kuchling)
Date: 23 Aug 1999 23:56:03 GMT
Subject: pyKDE
References: <rs2olmiv5u288@news.supernews.com>
Message-ID: <slrn7s3obr.gh.amk@207-172-112-26.s26.tnt4.ann.va.dialup.rcn.com>

On Mon, 23 Aug 1999 15:03:18 GMT, Jeremy <jeremy at exit109.com> wrote:
>Is anyone using pyKDE?  I'm trying to get it to work.  It compiled fine,
>but actually trying to use it, even just 'import kde', results in:
> /usr/local/lib/python1.5/lib-dynload/libkdecmodule.so: symbol
> className__C10QImageDrag: referenced symbol not found

Where are Qt and KDE installed?  Can Solaris find the relevant shared
libraries?  Try setting the LD_LIBRARY_PATH environment variable to
/opt/kde/lib, or wherever you've installed KDE.  

(I have pyKDE installed, and have written part of a KDE application,
but haven't gotten around to finishing it yet.  It's a data point that
pyKDE really does work, though.)

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
any-technology-indistinguishable-from-magic-is-too-mysterious-to-trust-ly y'rs
    -- Tim Peters, 16 Feb 1999



From tim_one at email.msn.com  Wed Aug  4 23:49:36 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 4 Aug 1999 23:49:36 -0400
Subject: Python microthreads
In-Reply-To: <FFy8ur.LHC@world.std.com>
Message-ID: <000301bedef5$89096280$662d2399@tim>

[Will Ware, on his microthreads vs Christian Tismer's Stackless Python]
> ...
> Yes, they're independent. Christian has already been in
> consultation with the gods of Python, so I think his work is
> more in the mainstream spirit of the Python codebase.

So far Guido has neither embraced nor rejected it; still a work in progress,
of uncertain outcome.  OTOH, your changes will never go in, because you went
out of your way to change ceval.c to use both brace and indentation styles
Guido hates <0.9 wink>.

> I don't think CPython loses anything by being stackless,

So far it has lost a little speed.  That can probably be regained.  Part of
Christian's agenda is seeing what it would take to implement 1st-class
continuations (in the Scheme sense) too, and his frame manipulations are
general enough to support that, and that costs something beyond mere
stacklessness.

> so maybe stacklessness could be folded into the core release.

Definitely the hoped-for outcome.

> So implementing microthreads separately will become an increasingly
> bad idea.

Yup.  You would gain some nice properties too; e.g., you could lose the
hardcoded limit on frames per thread, and run many more threads because the
control structure for each could be much smaller thanks to not preallocating
space proportional to that max # of frames.

> I don't know if Christian's version of eval_code2 can be
> stepped a controlled number of opcodes.

There's no direct provision for that now.

> That would need to be done for microthreads.

Python already bytecode-slices "real threads" every N opcodes (set via
sys.setcheckinterval); you probably didn't need to invent another mechanism
for that, and it's more efficient to hijack the one that's there (checking
two counters per eval-loop trip is twice the pure overhead of checking one
...).  So this one doesn't seem "deep".

Guido has let on that his primary objection to fake threads of any flavor is
that-- without an uncertain amount of additional widespread work --they'll
all block at I/O calls.  If you think threads are good only for overlapping
I/O (with other I/O or with computation), that's a killer objection.  I
personally see lots of uses for threads, but almost none for I/O <wink>.

doing-i/o-is-like-picking-someone-else's-nose-ly y'rs  - tim





From gutier at intergate.bc.ca  Thu Aug 19 01:19:34 1999
From: gutier at intergate.bc.ca (Gerald Gutierrez)
Date: 18 Aug 1999 22:19:34 -0700
Subject: Python-based PIM software?
Message-ID: <m3g11gwdnd.fsf@intergate.bc.ca>


Is anyone aware of any Python-based PIM software such as calendars and
addressbooks? They don't seem to exist ...

Thanks.

Please forward replies to gutier at intergate.bc.ca




From Oliphant.Travis at mayo.edu  Mon Aug  2 19:26:21 1999
From: Oliphant.Travis at mayo.edu (Travis Oliphant)
Date: Mon, 2 Aug 1999 18:26:21 -0500 (CDT)
Subject: Dislin is a nice plotting package
Message-ID: <Pine.LNX.4.10.9908021807550.12462-100000@us2.mayo.edu>

A users experience with plotting under Python:

I know plotting has been discussed on this list before.  Many people
indicate that the lack of quality plotting is preventing them from using
Numerical Python as a complete replacement for IDL or MATLAB.  

I have been using gist for the past several months and found it to be
excellent for investigative plotting (except I miss axes on surface
plots).  The plot quality and flexibility with gist is excellent (it is
particularly easy to use for me with the matlab-like plotting commands I
implemented) 

However, recently I needed to produce plots with legends on them so I
turned to DISLIN.  I had seen DISLIN before and even had it installed but
it looked pretty complicated to use so I hadn't tried it since I wasn't
desperate for good plots.  Well, I've now tried it and I'm impressed.
It is a COMPLETE plotting system.  After a bit of reading to understand
the plotting paradigm (it is a very powerful but non-obvious system) I can
now make any plot I need with the system, and configure it to my hearts
content.  

There are even quick plotting functions defined so that I can make  
surface plots (with axes) and color plots with an automatic color bar,
with a single plot command.  I can even use dislin and gist together which
works very well.

An object-oriented wrapper around the quick-plots so that all the
properties could be set easily would make dislin quite phenomenal for use
with Python. 

I don't mean this post to discourage work on the fine plotting efforts
underway with PIDDLE, GRAPHITE, and Snow.  These packages will only
improve the plotting choices under Python.  I just wanted to express my
pleasure at using DISLIN under Python and encourage others who may be
frustrated about their ability to produce quality output now or who
may be holding back from adopting Python as their interactive
analysis environment to check out DISLIN and see for yourself if it will
serve you.

-Travis

P.S. The dislin homepage is 

http://www.linmpi.mpg.de/dislin




From fredrik at pythonware.com  Tue Aug 31 03:00:13 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 31 Aug 1999 07:00:13 GMT
Subject: Grids in Tkinter
References: <936050813snz@vision25.demon.co.uk>
Message-ID: <011801bef37e$bd171cf0$f29b12c2@secret.pythonware.com>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
> Can anyone point me to anywhere that documents how Grids work in
> Tkinter? A short example program would be nice.

http://www.pythonware.com/library/tkinter/introduction/dialog-windows.htm#AEN1406
http://www.pythonware.com/library/tkinter/introduction/grid.htm

</F>




From guido at cnri.reston.va.us  Fri Aug 20 10:44:10 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 20 Aug 1999 10:44:10 -0400
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk> <37BC8D05.142E1DB0@aw.sgi.com> <37BC9CE8.67097FC6@vision.arc.nasa.gov> <37BCD6CF.B7A186BC@montana.com> <5lzozmskm3.fsf@eric.cnri.reston.va.us> <00dd01beeb11$e41a68e0$f29b12c2@secret.pythonware.com>
Message-ID: <5lso5ese9x.fsf@eric.cnri.reston.va.us>

"Fredrik Lundh" <fredrik at pythonware.com> writes:

> Guido van Rossum <guido at cnri.reston.va.us> wrote:
> > Actually, the accepted idiom (all over the place in code I write) is
> > 
> >         toplevel_window.protocol("WM_DELETE_WINDOW", close_callback)
> > 
> > The close_callback function takes no arguments.  In simple cases, you
> > can use
> > 
> > root.protocol("WM_DELETE_WINDOW", root.destroy)
> > 
> > or
> > 
> > root.protocol("WM_DELETE_WINDOW", root.quit)
> 
> how about making the destroy variant the
> default?

Good idea.  I've thought of this before, and somehow never got to it.

I can't see any reason not to do it; I believe this is the default in
Tk anyway, but Tkinter's destroy() does a little more than Tk's
internal destroy.  Unless I hear a clamor of complaints I'll go ahead
and do this.

--Guido van Rossum (home page: http://www.python.org/~guido/)



From wpeterson at cfa.harvard.edu  Mon Aug  9 12:03:04 1999
From: wpeterson at cfa.harvard.edu (William Peterson)
Date: Mon, 09 Aug 1999 12:03:04 -0400
Subject: Tkinter -- tarred up documentation?
References: <934062911snz@vision25.demon.co.uk>
Message-ID: <37AEFBB8.DB9947D9@cfa.harvard.edu>

The Python distribution contains a wonderful utility, websucker, that's
handy for this task. Look in the tools subdirectory.

Bill

Phil Hunt wrote:

> On the Pythonware site (www.pythonware.com) there's a some Tkinter
> documentation entitled "Introduction to Tkinter", and "Tkinter
> Class Reference pages".
>
> Unfortunately these are stored as single HTML pages. Is there anywhere
> where they are tarred and gzipped up, so I can easily download them
> to my PC?
>
> --
> Phil Hunt....philh at vision25.demon.co.uk




From mlh at idt.ntnu.no  Thu Aug 26 11:11:49 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 26 Aug 1999 17:11:49 +0200
Subject: Is there a pattern...
Message-ID: <y0jg116vaoa.fsf@vier.idi.ntnu.no>

Hi!

While checking some codee (that I had not written myself - and it was
in Java, not Python...) I encountered the following situation...

The code included classes and methods for creating graphs (i.e.
networks), in quite a nice manner. Each vertex had a list of edges,
and the edges themselve took care of registering themselves (and
removing themselves) with the vertices. Thus, you could make a simple
graph like this:

x = Vertex("some value")
y = Vertex("some other value")
Edge(x,y)

Now - if you want to remove the edge, you would do something like:

x.getOutEdge(0).deleteMe()

That is - you would get a hold of the edge you wanted to remove, and
call its delete method. (In Python maybe you could use del, but that's
beside the point). Now - this method called corresponding methods in
each of its two vertices (in this case x an y) that actually removed
it from their lists of edges. This seems OK - *except* that these
methods should *only* be called from the edge and nowhere else...

So I started to wonder - is there an intelligent design pattern that
addresses this problem, i.e. methods that should only be called from a
specific class or a specific set of objects or something? (Somewhat
like friend classes in C++, I guess - if I don't remember it all
wrong...)

Of course, you *could* just appeal to the programmers conscience and
tell him not to call the methods anywhere else, but...

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From chad at vision.arc.nasa.gov  Thu Aug 19 19:47:09 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Thu, 19 Aug 1999 23:47:09 +0000
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk>
Message-ID: <37BC977D.9699419B@vision.arc.nasa.gov>

Stuart Reynolds wrote:

> Does anyone know how to tell if Tk has already instaniated a Toplevel
> window? I want to be able to open several Toplevel windows in any order.

Try this:

from Tkinter import *

def makeNewWindow():
    from Tkinter import _default_root
    if _default_root is not None:
        master=Toplevel()
    else:
        master = None
    f = Frame(master=master)
    b = Button(master=f, text="Hello")
    b.pack(side=LEFT, fill=BOTH)
    Pack.config(f)

makeNewWindow()
makeNewWindow()
makeNewWindow()

NOTE - The "from Tkinter import _default_root" can be avoided by importing
Tkinter directly at the beginning (ie. "import Tkinter"), and accessing it
through
the Tkinter module.  Otherwise, the _default_root reference you have won't get

updated when Tkinter changes it.





From kernr at mail.ncifcrf.gov  Thu Aug 19 18:28:03 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Thu, 19 Aug 1999 22:28:03 GMT
Subject: No man pages? (was Re: More random python observations from a perl programmer)
References: <199908192155.PAA08720@jhereg.perl.com>
Message-ID: <37bc80ff.23592487@news.erols.com>

On Thu, 19 Aug 1999 15:55:47 -0600, Tom Christiansen
<tchrist at jhereg.perl.com> wrote:

[snip]

>% (echo 'import os' ; echo 'print os.execv.__doc__' ) | python 

You can simplify that to 
% python -c 'import os; print os.execv.__doc__'
:-)

>Execute an executable path with arguments, replacing current process.
>
>        path: path of executable file
>        args: tuple or list of strings
>
>Well, that's better than nothing.
>
>% (echo 'import os' ; echo 'print os.__doc__' ) | python 
>None
>
>But that isn't. :-)

True.  Missing doc-strings has been the perennial biggest problem with
Python's documentation.

>I'd guess for a module, you should traverse its members and
>print docs for all that have them.
>
>I'm still getting ideas for autogenerating manpages. :-)

We've probably thought of them, too (well, the autogeneration part at
least).  The biggest problem with using them on third-party modules is
that there is no consistent standard for them like pod.  We have
Guido's recommendations from his style guide, and we have things like
Structured Text, but we also have have a zillion other styles for
docstrings.  We could probably enforce a standard for the standard
library, though.  Use Deja.com to read a thread named "Pod" we had
here back in October and November of 1998 where we talked about most
of these things.  But ...

>But I haven't looked at the HTML or latex docs yet.  Probably
>won't till after the conference or so.

... more of the information you would want in a manpage is probably in
the latex sources (and, by extension, HTML) than in the docstrings.

Things will get even easier when the conversion to XML or SGML is made
(RSN).

>--tom

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From nhead at houbits.com  Mon Aug  9 09:07:11 1999
From: nhead at houbits.com (Nigel Head)
Date: Mon, 09 Aug 1999 15:07:11 +0200
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com> <37AA804D.9A1317C0@houbits.com> <37AEA5A6.5C556643@houbits.com> <87907l5eip.fsf@psyche.evansnet>
Message-ID: <37AED27F.1C3D7CC9@houbits.com>


Carey Evans wrote:

> If I decide that $100 is a reasonable copying fee, and whoever I sell
> it to agrees or is convinced by my marketing, I can "sell" it for
> $100.  (See the definitions.  Note that I don't have to justify to you
> what I sell it for.)

No probs - you're selling my package, with my name on it!  Credit and future
enquiries come to me (possibly!).

> Or I can use it in a $1000 package and sell
> that, under section 8, as long as it isn't obvious I'm using your
> software!

Also no problems with that from my side.

I guess what would really bug me is for people to hijack the package, remove
original credits and resell straight with no added value of their own. I
think that should be disallowed; that's all!

Nigel.




From bwarsaw at python.org  Thu Aug 19 11:38:09 1999
From: bwarsaw at python.org (bwarsaw at python.org)
Date: Thu, 19 Aug 1999 11:38:09 -0400 (EDT)
Subject: PIGgies?
References: <19990818171523.A22207@digex.net>
Message-ID: <14268.9441.831503.31111@anthem.cnri.reston.va.us>

>>>>> "JL" == John Landahl <jpl at digex.net.nospamplease> writes:

    JL> These are my opinions, of course, and I graciously solicit the
    JL> opinions of others.  What do you think of the term "PIGgies"?

As for the term PIGgies, I always called it a "PUG" until I heard
about BayPIGgies started.  I don't know what a PUG is (other than a
pugnose), and a PIG has this connotation of being swallowed whole by
the Python.  I have no idea what I'm actually talking about.
    
    JL> Do you feel that there is a need (and room) for separate
    JL> interest and advocacy groups?  Do either the Python Software
    JL> Activity or the Python Consortium already fill the advocacy
    JL> role?

Anyway, I definitely think there is use in having informal local user
group chapters.  Think of them as social gatherings with a distinctive
geek bent.  Plus it's fun to meet your Pythonic neighbors, hobnob,
eat, and learn or discuss a few things.  So I encourage other regions
to start up PUGs/PIGgies.  If enough of them start up and have their
own Web pages, I'd even be happy to add something to python.org that
points to them.

-Barry



From billtut at microsoft.com  Thu Aug  5 23:12:52 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Fri, 6 Aug 1999 03:12:52 GMT
Subject: Any clues to source of this delay?
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB91039@RED-MSG-50>

Windows does indeed support this socket option.
Here's what MSDN has to say about the option (in order to fill in all of the
icky details):

TCP_NODELAY 
The TCP_NODELAY option is specific to TCP/IP service providers. The Nagle
algorithm is disabled if the TCP_NODELAY option is enabled (and vice versa).
The process involves buffering send data when there is unacknowledged data
already in flight or buffering send data until a full-size packet can be
sent. It is highly recommended that TCP/IP service providers enable the
Nagle Algorithm by default, and for the vast majority of application
protocols the Nagle Algorithm can deliver significant performance
enhancements. However, for some applications this algorithm can impede
performance, and TCP_NODELAY can be used to turn it off. These are
applications where many small messages are sent, and the time delays between
the messages are maintained. Application writers should not set TCP_NODELAY
unless the impact of doing so is well-understood and desired because setting
TCP_NODELAY can have a significant negative impact on network and
application performance. 

Bill


> -----Original Message-----
> From: Bruce Dodson [mailto:bruce_dodson at bigfoot.com]
> 
> 
> This isn't unique to Python.  It has to do with the way TCP 
> buffers data
> within a time window by default.  The data is sent when some 
> threshold size
> is surpassed or when the time window expires (and I guess you saw that
> sending zero bytes can force it to flush on some platforms).  For many
> applications this buffering gives a performance improvement, 
> but in your
> case (the kind of application that wants to do lots 
> round-trips per second
> with small amounts of data in each request), it kills performance.
> 
> To optimize a socket for this kind of conversation, you can use socket
> option TCP_NODELAY, but this is not a portable solution; I 
> think it's a bit
> different on Windows than on Linux, and it is not available on all
> platforms.  I found no reference to TCP_NODELAY in the Python 
> reference,
> which is probably for the best given its non-portability.  If you are
> targetting a platform that supports it, you can probably do 
> something like:
> 
>   if (os.name == 'nt'):
>     TCP_NODELAY = 1
>     theSocket.setsockopt(IPPROTO_TCP, TCP_NODELAY, 1)
> 
> Depending on the nature of the conversation, one or both sides of your
> connection may want to do that.  Protecting the nonstandard 
> NODELAY option
> with a platform test means the code will still work on other 
> platforms,
> although it will still have a "glass ceiling" on platforms 
> that you haven't
> planned for.
> 
> Bruce
> 
> 



From kperacles at geocities.com  Sun Aug  1 16:15:11 1999
From: kperacles at geocities.com (Karl Putland)
Date: Sun, 1 Aug 1999 14:15:11 -0600
Subject: spawnv
References: <19990801171952.JFVG5731@default>
Message-ID: <h_1p3.2850$424.327844@news.uswest.net>

This worked for me on a project I did a couple of months ago.  It did its thing for a couple of minutes then when it exited the Python script resumed.

...
import os
...
    os.spawnv(os.P_WAIT, os.path.join(gmpath,'gmw4.exe'),
              [r' /u:MASTER', r' /p:ACCESS', r' /r:69,5'])
...

spawnv (mode, path, args) 
Execute the program path in a new process, passing the arguments specified in args as command-line parameters. args may be a list or a tuple. mode is a magic operational constant. See the Visual C++ Runtime Library documentation for further information. Availability: Windows. New in version 1.5.2. 

P_WAIT 
P_NOWAIT
P_NOWAITO
P_OVERLAY
P_DETACH
Possible values for the mode parameter to spawnv() and spawnve(). Availability: Windows. New in version 1.5.2. 


Karl Putland
kperacles at geocities.com



Craig Curtis <c.curtis at worldnet.att.net> wrote in message news:19990801171952.JFVG5731 at default...
> I am trying to spawn c:\windows\notepad.exe from a Python program.  I want
> the python script to wait for Notepad to exit.  
> How can I do this?  
> 
> I have tried all combinations I can think of using execv, spawn, system and
> cannot get it to work.
> 
> Could someone please post the exact command sequence to type in.
> 
> Thanks in advance,
> Craig Curtis
> 
> 




From skaller at maxtal.com.au  Tue Aug 17 00:03:50 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Tue, 17 Aug 1999 04:03:50 GMT
Subject: C's isprint() concept?
References: <000201bee78f$6bbf6ec0$f22d2399@tim>
Message-ID: <37b8def1.474083836@news.triode.net.au>

On Sun, 15 Aug 1999 22:31:18 -0400, "Tim Peters" <tim_one at email.msn.com> wrote:

>
>Take a look at string.translate.

	And avoid it, since it isn't compatible with unicode or UTF-8
encodings :-(

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From kaz at ashi.FootPrints.net  Tue Aug 24 02:14:25 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Tue, 24 Aug 1999 06:14:25 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk>  <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu>  <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu>  <37c126a6$0$813@newsreader.alink.net>  <slrn7s2dtk.vsl.kaz@ashi.FootPrints.net> <37c21fae$0$216@nntp1.ba.best.com>
Message-ID: <slrn7s4e2u.hkm.kaz@ashi.FootPrints.net>

On Mon, 23 Aug 1999 21:29:34 -0700, Xah <xah at best.com> wrote:
>I gather you are a newbie. Normally I'd just kick newbie's ass for the fun
>of it, but this time I feel magnanimous and will give you the benefit of a
>doubt.

Usenet can be entertaining, but kooks bore me.

So I guess it's that time when we sadly part with that sound that tears make
when they hit the keyboard.

It sounds a little bit like... *plonk*



From kzaragoza at mediaone.net  Thu Aug 26 13:07:43 1999
From: kzaragoza at mediaone.net (Kris J. Zaragoza)
Date: Thu, 26 Aug 1999 17:07:43 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net>
Message-ID: <slrn7sasvb.397.kzaragoza@kzaragoza.ne.mediaone.net>

On Thu, 26 Aug 1999 14:33:40 GMT, Paul Prescod <paul at prescod.net> wrote:
>I think it would be more pythonish if this would return a runtime (or
>even compiletime) error:
>
>def foo():
>	a=5
>
>b=foo()
>
>It shouldn't just return None. That's a source of errors.

The question here is, where is the error?  Unlike Basic (VB,
LotusScript, etc.), Python does not differentiate between blocks that
return a value (a Function), and those that do not (a Sub).  This
means that the above function foo() is, as defined, syntactically
correct.

Now when the call is made, you need to make a determination as to
whether the calling line is in error, or the function itself is in
error.  With your example, it could be either.  The author of the
function may have erred in neglecting to insert a return statement at
the end of the function.  If, on the other hand, the function does not 
intend to return any values, then the author of the calling line made
the error in simply assuming that it would.

I tend to like Python's current way of handling things.  If a function 
does not explicitly return a value, None is returned instead.  It's a
clean solution to an otherwise intractable problem.

How do you propose fixing this in Python 2?  Would you enforce the
separation of blocks that do and don't return values?  Would you
include some sort of pragmas that tell the compiler and runtime that a 
value is not to be returned?  Other ideas?  (Personally, I wouldn't
mind if this stayed as is.  I just read the documentation or doc
string to see if a given function is meant to return a value.)

-Kris

-- 
Kris J. Zaragoza           | "Unfortunately, most people can't out-think a
kzaragoza at mediaone.net     | grapefruit."  --Jon Bodner



From python-list at teleo.net  Sun Aug  1 15:54:20 1999
From: python-list at teleo.net (Patrick Phalen)
Date: Sun, 1 Aug 1999 12:54:20 -0700
Subject: RMI and Python
References: <Pine.GSO.4.10.9908011131490.25758-100000@xmission.xmission.com>
Message-ID: <99080113023108.02407@quadra.teleo.net>

On Sun, 01 Aug 1999, Tim Egbert wrote:
> Java is able to do remote method invocation (RMI) which is a neat concept.
> It is the basis for Jini, another neat idea.  Does Python have anything
> like RMI?

If you can forego the concept of agents, take a look at

www.xmlrpc.com

and

www.pythonware.com/products/xmlrpc

Zope, the Python-based application server platform (www.zope.org), now
includes XML-RPC capability in the core




From dhandly at cup.hp.com  Tue Aug  3 03:50:52 1999
From: dhandly at cup.hp.com (Dennis Handly)
Date: 3 Aug 1999 07:50:52 GMT
Subject: HELP: ImportError on HPUX
References: <7o4c1s$n7d$1@nnrp1.deja.com>
Message-ID: <7o670s$kk1$3@ocean.cup.hp.com>

rceci at adams.com wrote:
: I found an old reference to C++ modules in HP needing
: "Python/importdl.c" to have shl_load changed to cxxshl_load and what
: not.  I am not sure that is still valid,
: HP aC++ B3910B A.01.18

No, that is only for cfront.

: /usr/lib/dld.sl: Unresolved symbol: __StaticCtorTable_Start (data)
: /usr/lib/dld.sl: Unresolved symbol: __ZStaticCtorTable_End (data)
: /usr/lib/dld.sl: Unresolved symbol: __StaticCtorTable_End (data)
: /usr/lib/dld.sl: Unresolved symbol: __ZStaticCtorTable_Start (data)

These symbols must be in your a.out.  See the URL below for some stubs
to workaround this.

: /usr/lib/dld.sl: Unresolved symbol: U_update_state_vector (code)
: /usr/lib/dld.sl: Unresolved symbol: U_get_shLib_unw_tbl (code)
: /usr/lib/dld.sl: Unresolved symbol: U_resume_execution (code)
: /usr/lib/dld.sl: Unresolved symbol: U_get_previous_frame_x (code)
: /usr/lib/dld.sl: Unresolved symbol: U_get_unwind_table (code)
: /usr/lib/dld.sl: Unresolved symbol: U_get_shLib_text_addr (code)
: /usr/lib/dld.sl: Unresolved symbol: U_get_unwind_entry (code)

These symbols are in /usr/lib/libcl.sl and must be attached to your a.out.

http://docs.hp.com/hpux/content/aCC/guide45/distributing.htm



From brianr at dnr.state.ak.us  Thu Aug  5 14:32:03 1999
From: brianr at dnr.state.ak.us (Brian Raynes)
Date: Thu, 05 Aug 1999 10:32:03 -0800
Subject: Python and Red Hat Linux 6.0
References: <379CCAC0.AD5D1C2F@postoffice.pacbell.net>
Message-ID: <37A9D8A3.D541468F@dnr.state.ak.us>


Rob wrote:
> 
> When programming in Perl or Python, I am using a header like the
> following on an executable file to cause the shell (bash or csh) to run
> the scripts as Perl/Python files:
> #!/usr/bin/perl   for Perl   and #!/usr/bin/python   for Python. I've
> checked and the compilers are located in the appropriate directories. I
> keep getting this when I run scripts:
> bash: <script name>: command not found. Does Linux support the '#!'
> notation in the file header, or do I always have to run the scripts by
> explicitly typing in 'perl' or 'python' before the script name?


I also recently had a similar problem, I had created the
files using Wordpad and saved them to the Linux server.  I
thought I was using a UNIX-safe text format, but I later
discovered it was in DOS text format.  Just another idea.



From roy at popmail.med.nyu.edu  Sun Aug 22 11:06:41 1999
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Sun, 22 Aug 1999 11:06:41 -0400
Subject: More random python observations from a perl programmer
References: <000001beec6d$9753e240$542d2399@tim> <m2iu677ujv.fsf@netmeg.net> <roy-2208991100070001@mc-as01-p27.med.nyu.edu>
Message-ID: <roy-2208991106420001@mc-as01-p27.med.nyu.edu>

Arrrgh!  I should proofread more carefully before I post.

I wrote:

> Roof_material and roof_color are NOT orthogonal; while color can be white
> or green, it can only be so if material is not ceramic_tile.

Obviously, I meant white or blue.

> x = 4 + 3
> x = 4.0 + 3.0
> x = "Four" + "Three"
> 
> are legal, with x getting the value 4, 4.0, and "FourThree" respectively. 

Make that 7, 7.0, and "FourThree".  It's been a tough week, but I think I
should still be able to handle single-digit integer addition on my own :-(



From clgonsal at keeshah.penguinpowered.com  Sat Aug 28 17:15:59 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Sat, 28 Aug 1999 21:15:59 GMT
Subject: Converting from perl
References: <7pujub$75v$1@nnrp1.deja.com> <m3ogfwrfwo.fsf@atrus.jesus.cam.ac.uk> <7pve6j$qng$1@nnrp1.deja.com>
Message-ID: <slrn7sgkc2.4ee.clgonsal@keeshah.penguinpowered.com>

On Wed, 25 Aug 1999 00:43:01 GMT, sp00fD <sp00fD at yahoo.com> wrote:
> In article <m3ogfwrfwo.fsf at atrus.jesus.cam.ac.uk>,
>   Michael Hudson <mwh21 at cam.ac.uk> wrote:
> > sp00fD <sp00fD at yahoo.com> writes:
> >
> > To be honest, I don't see a lot of value in wrapping this up in a
> > class. For a straight through shell scripty type job like this I
> > generally don't bother. Python is not about ramming methodologies down
> > your throat.
> >
> 
> Honestly, me neither, more than anything I wanted to get a feel for
> them, basically I was bound and determined to create a class, but when I
> was done, it was obviously stupid.  Unfortunately, most of the stuff I
> do is "straight through shell scripty" so, hopefully I can figure out
> some decent way to use the OO features of python for that.

Well, one thing I would probably avoid in a class is having any
references to "global" sorts of things like sys.argv. Classes are best
used to package reusable chunks of code. If you go and parse command
line arguments in the class, it's hard to use that class for anything
else. The other great thing about classes is polymorphism, but you
pretty well need more than one class to make that happen...

Since this particular chunk of code is almost entirely command line
argument parsing, I don't see how you can make use of classes in a very
effective way.

I do understand that you're doing this as a learning excercise though,
so to give you an idea of how OO might be useful for a very similar
problem to the one you have...

Suppose that some of your machines don't work with rcp. Suppose some
needed ssh, others FTP, others HTTP-put. Maybe in some of them you store
everything in a database, and you need to do sql statements. The end
user doesn't care. They just want to be able to say "get my file over
there". You could do something like this:

    import sys
    import os
    from ftplib import FTP

    class DestinationMachine:
        pass # sometimes I wish Python had abstract classes or interfaces...

    class RCPDestinationMachine(DesitinationMachine):
        def __init__( self, hostname ):
            self.hostname = hostname

        def copy( self, src, dest ):
            if os.system( "rcp %s %s:%s" % (src, self.hostname, dest) ):
                raise "Argh!" # this is pretty cheesy...

    class FTPDestinationMachine(DesitinationMachine):
        def __init__( self, hostname, userid, passwd ):
            self.hostname = hostname
            self._userid = userid
            self._passwd = passwd

        # warning: totally untested code ahead
        def copy( self, src, dest ):
            server = FTP( self.hostname, self._userid, self._passwd )
            fp = open(src)
            server.storbinary( "STOR " + dest, fpi, 4096 )
            fp.close()
            server.quit()

    # do the same for other copying mechanisms

Then you could make a map that maps server names to the appropriate
DestinationMachine object for that server. 

    tempMachineList = (
        RCPDestination( "frodo" ),
        RCPDestination( "bilbo" ),
        FTPDestination( "jupiter", "root",      "zeus" ),
        FTPDestination( "saturn",  "foofoo",    "barbar" ),
        FTPDestination( "mars",    "plaintext", "password" )
    )
    hostMap = {}
    for host in tempMachineList:
        hostMap[host.hostname] = host
    del tempMachineList
    # now hostMap maps from hostnames to DestinationMachine objects

Your copy method (in your existing class) would then look something like:

        def copy(self):
                _stuff = self.files()

                try:
                        for hostname in self.get_hosts():
                                print "Sending %s to %s %s" % \
                                (_stuff[0], host, _stuff[1])
                    
                                hostMap[hostname].copy( src, dest )
                except:
                        print "Error during copy - oops"

and it would automagically use rcp, ftp, or whatever machanism was
appropriate for the destination machine.

You could then also reuse these classes in other scripts that needed to
copy files between machines as part of their larger duties. You could
even put the code that creates the mapping and all of the
'DestinationMachine' objects into a module so that when you
get/find-out-about a new machine, all of your scripts will know about it.

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From thomas at xs4all.nl  Mon Aug 16 11:15:06 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Mon, 16 Aug 1999 17:15:06 +0200
Subject: Difficulty configuring web server for Python
In-Reply-To: <7p9855$5tg$1@nnrp1.deja.com>; from mbf2y@my-deja.com on Mon, Aug 16, 1999 at 02:44:56PM +0000
References: <7p9855$5tg$1@nnrp1.deja.com>
Message-ID: <19990816171506.K21639@xs4all.nl>

On Mon, Aug 16, 1999 at 02:44:56PM +0000, mbf2y at my-deja.com wrote:

> I built a simple CGI script and kept getting error 500, so I simplified
> to the below script:

> #!/usr/home/campusch/usr/local/bin/python

Does your webserver do something like chroot() ? If so, is this the correct
path according to the webserver ?

> print "Content-type: text/html"
> print
> print "<HTML><BODY><H1>Hello World!</H1></BODY></HTML>"
> print

> 4) python itself has been chmod ugo+x so the web server process should
> be able to run it.

And all the directories below it are chmod <webserver>+x as well ?

> 5) This script has been chmod ugo+x so the web server process should be
> able to run it.

Has it also been set readable ? it needs to be readable for the webserver to
start python to interpret.

> [Sun Aug 15 21:59:31 1999] [error] (2)No such file or directory: exec of
> /usr/local/etc/httpd/cgi-bin/simp.py failed
> [Sun Aug 15 21:59:31 1999] [error] Premature end of script headers:
> /usr/local/etc/httpd/cgi-bin/simp.py

> While the "no such file or directory" threw me, this is not a file not
> found error... when I tried to run a script that really wasn't there, I
> got:

The 'no such file' probably refers to the script interpreter, not the script
itself... Is the script in unix text format, or DOS ? 'DOS text format'
means line endings are marked using '\r\n', UNIX uses just '\n's. Having the
file in DOS text would mean the webserver is looking for
'/usr/home/campusch/usr/local/bin/python^M'. You either need to run dos2unix
on those files (or dos2bsd as it's called under BSDI) or add a (sym)link
from python to python^M. (To give a hint... using the tcsh shell you do it
with

ln -s python python<control-v control-j>

in the /usr/home/campusch/usr/local/bin directory.

G'luck :)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From tim_one at email.msn.com  Sun Aug 15 22:31:24 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 15 Aug 1999 22:31:24 -0400
Subject: Compound Assignment Operators ( +=, *=, etc...)
In-Reply-To: <37B675A9.53C2BBE0@getaway.net>
Message-ID: <000501bee78f$6f0523e0$f22d2399@tim>

>> ... does
>>
>>     z = x
>>     x += y
>>     print z
>>
>> mean
>>
>>     x = x + y   # x is now bound to an entirely different object
>>     # and the old x is printed
>>
>> or
>>
>>    x.extend(y)   # x is still the same object, but mutated
>>    # and the new x is printed
>>
>> ?  The resulting *values* compare equal either way, but there are
>> differences in efficiency and effects on object identity.

[Andrew McDowell]
> I don't see where the "non-obviousness" comes into play...
> (maybe that's why it's not obvious <wink>)

In that reasonable people can (& do) expect either behavior.

> Why would you confuse x += y (which would seem to imply the
> former) to mean x.extend(y) ?

Efficiency; and, to half the reasonable people in the world, obviousness
<wink>.

> It _is_ and assignment statement after all..

That's my mild preference, and it's easier to implement that way; but some
reasonable people want it the other way.  Consider the NumPy extension,
where "x" can denote a matrix with millions of elements and consuming many
Mb of storage.  There you may desperately not want e.g.

    x += 1

to create an entirely new similar-sized matrix, yet still want the economy
of expression.  Guido's preferred scheme caters to either behavior.

> Not to touch on a sore topic, but hop over to comp.lang.perl and
> check out the (rather long and heated) comparison of perl and python.

So *that's* what it is -- I thought it was a flame war.

> The lack of these statements is causing quite a bit of a stir.

I expect you could post "have a nice day!" to that thread and also "cause"
quite a bit of a stir.  Flame wars aren't "about" the topics being toasted,
y'know <0.1 wink>.

>> ... it's up to x.__addeq__(y) to decide what "+=" (& so on) means.

> Why addeq??  Why not just add?

So that it's *possible* for the implementation of x's type to arrange for

    y = x + 1

to create a new object but for

    x += 1

not to.  If you don't want to allow for mutation, __addeq__ isn't needed
(and you can find an implementation sketch in DejaNews for this case -- it's
easy to implement in Python's front end).

lettuce-should-never-be-chewed-twice-ly y'rs  - tim





From jstevens at basho.fc.hp.com  Fri Aug 13 23:06:43 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:06:43 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132013.OAA03423@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:13 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cf33@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :And cost a lot, big time, in the real world also.  The most common
> :lament from a section manager: "I've got all this Perl code that
> :<fill-in-a-name> wrote, and nobody else can understand this stuff!"
> 
> I am sick and tired of the implication that because stupid people write
> stupid Perl, that Perl is stupid.

That wasn't my implication at all.

I thought I was being quite clear.  Perl allows for a greater range
in expression, which allows for a greater amount of ambiguitiy,
which creates a situation where maintaining code costs much, much
more.

> Stupid people are associated with
> everything!  Perl is also 100x more popular than Python -- simple because
> even stupid people *are able* to use it!

Your premise and your conclusion are not logically relatable.

Proof, please, that Perl is 100x more popular than Python?
(To prove this, you must find a random sample of people who
have used both, then prove that of those who use both equally,
that 100x as many prefer Perl).

Proof, please, that the reason Perl is more popular than Python
is due to the fact that stupid people are able to use it, while
stupid people are *NOT* able to use Python?

> And they do.  And this is what
> we get for making something that even stupid people can use.  I hope that
> Python should be someday cursed with one tenth the number of stupid people
> who now write Perl.  Then your newsgroup can be as messy as this one.
> 
>     1.  Assume competence.

I do.  I have noted, though, that competent people are not the same
as disciplined people.  I have noted that competent people are not
always competent at everything.  I have even noted that brilliance,
and the ability to created unmaintainable code are not always
mutually exclusive.

>     2.  Train the ignorant.

Of course.  For five years, I did just that.

>     3.  Fire the stupid.

Harsh, but it works.

John S.




From robin.boerdijk at nl.origin-it.com  Wed Aug  4 13:45:38 1999
From: robin.boerdijk at nl.origin-it.com (Robin Boerdijk)
Date: Wed, 4 Aug 1999 19:45:38 +0200 
Subject: Any clues to source of this delay?
References: <al8vhaxtg1y.fsf@sirppi.helsinki.fi> <14247.22099.356207.503879@dolphin.mojam.com> <al8d7x4qkoc.fsf@sirppi.helsinki.fi>
Message-ID: <0E16861EE7BCD111BE9400805FE6841F0AB89844@c1s5x001.cor.srvfarm.origin-it.com>

> Oh, sorry, wasn't being explicit; it isn't slow start, as the ceiling is
on
> _single_ connection, however long it stays up. By opening multiple
> connections, the server's total handling capacity goes up but single
> connections' speed stays constant.

At the risk of being completely wrong: your problem might be caused by the
fact that TCP has a sliding window of 64Kb. This means that there can never
be more than 64Kb of outstanding, unacknowledged data on a single TCP
connection. When TCP reaches that limit, it will stop sending data until an
acknowledgement from the receiver makes the window 'slide' to the right. The
larger the size of the packets you sent, the sooner the windows fills up and
the less rountrips per second you will get.

Hoping this guess is not totally off,

Robin.






From ScherBi at BAM.com  Fri Aug 20 15:00:34 1999
From: ScherBi at BAM.com (ScherBi at BAM.com)
Date: Fri, 20 Aug 1999 15:00:34 -0400
Subject: Python+Apache+Oracle+???Persistent database connection???
Message-ID: <3F8BA119B48AD11197470008C7A493CB98B51D@HQOBGEX1>

Philip - 

You can acheive persistant database connectivity by converting your cgi apps
to pcgi.  pcgi (persistant cgi), was developed by Jeff Bauer, and is
available as a component of Zope (http://www.zope.org).  You can use pcgi
with or without Zope.

Good luck.

Bill

-----Original Message-----
From: Philip Payne [mailto:pnpayne at swissonline.ch]
Sent: Friday, August 20, 1999 2:50 PM
To: python-list at python.org
Subject: Python+Apache+Oracle+???Persistent database connection???


Hi,

Can any of you Python gurus tell me if the following would be
easy/possible/impossible and perhaps point me in the right direction???

I have Python (1.5.1), Apache (1.3.6), Oracle (8.0.5), PyApache (4.16),
and DCOracle (Python interface to Oracle - 1.2.1) on my Linux  box
(Caldera OpenLinux 2.2). It works fine.

But what I would like to have is a persistent database connection for
each Apache child server that is accessible to every Python script that
is fired up by that server. This because logging on to Oracle is VERY
expensive compared to simple database accesses. As far as I can see this
isn't possible with PyApache/DCOracle at present.

Now my Python level is middling, I have no experience in embedding or
extending Python, or in writing Apache modules, and my C is *VERY*
rusty, but from browsing some of the documentation and code this is my
limited understanding of what happens every time when Apache calls
PyApache to run a Python script:
        Py_NewInterpreter
        execute Python script
        Py_EndInterpreter

Logging on to Oracle in DCOracle returns a 'Connection' object. Am I
right in thinking that this object and every other Python object
automatically and ineluctably get destroyed by the Py_EndInterpreter
call??? If yes, is there any way to save it away 'somewhere' and get it
back later 'somehow' after the next Py_NewInterpreter call??? Or would
this require some sort of coordination between the PyApache and DCOracle
code??? Or is it impossible???

Any help gratefully appreciated.

Regards,
Philip Payne

-- 
http://www.python.org/mailman/listinfo/python-list



From mnot at mnot.net  Tue Aug 24 08:58:05 1999
From: mnot at mnot.net (Mark Nottingham)
Date: Tue, 24 Aug 1999 22:58:05 +1000
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan> <slrn7s408d.gvn.wtanksle@dolphin.openprojects.net> <hhvha5tswc.fsf@forcix.roof.lan>
Message-ID: <0cb801beee30$4e723120$0301a8c0@mnot.net>

I understand your frustration, but the inconveniece of adding a pass is
slight. If such a basic semantic feature of the language were optional,
things, IMHO, would go to hell in a handbasket. One of the major pluses of
Python over Perl in my book is the *lack* of being able to do everything
fourteen different ways; it confuses other people who have to read your
code.

> if foo:
>    # bar
>    # baz
> qux
>
> E.g. if you're developing code and just commented out a (not yet
> worked out branch) to work on a different part of the program.
> Requireing pass here is unecessary.
>
> I didn't ask for pass to vanish completely, i merely asked for it
> to optional instead of mandatory.






From m.faassen at vet.uu.nl  Mon Aug  9 05:13:30 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 9 Aug 1999 09:13:30 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com> <37AA804D.9A1317C0@houbits.com> <934115482snz@vision25.demon.co.uk> <37ae686e.59511102@news.ricochet.net>
Message-ID: <7om63q$cqc$2@newshost.accu.uu.nl>

phil <phil at ricochet.net> wrote:
[snip discussion of licenses, GPL, etc]
> I can't forbid commercial use. This was why i chose Python. I have no
> idea if people can or will  make money on my interface. If Python
> imposes commercial limitations then i'll have to drop it, pissed-ly,
> since i spent quite a bit of time writing an extension, but that's the
> way it goes.

I'm confused. Perhaps you're under the impression Python is distributed
under the GPL? It's not; it's distributed under its own license which
(read it) basically allows you to do anything you like with it, commercial
or not.

The GPL allows any use and modifications, as long as you supply the source
code of your modifications, also GPL-ed. And if you base your work on a
GPL-ed library the intent of the GPL is that you GPL your work as well.
By this means GPL-ed code tries to 'infect' anything that uses it (note that
the output of GPL-ed software can still be used freely, such as in the case
of the gcc compiler). The 'LGPL' (library GPL or these days 'lesser GPL')
is more liberal in the sense that non GPL code can use it without having to
be GPL itself.

All this is however not applicable to Python (though it may be to some
extension libraries), as Python does not come with a GPL license.

There are advantages to the GPL and the non GPL ways to open source code;
the GPL tends to safeguard against code splits and enhancements not making
it to the general public, while the BSD-style licenses (is the Python
license one in the practical usage sense?) allow practically any use, which
allows some enhancements to remain private, but also makes it easier for
companies to work with Python; companies which may eventually contribute
a lot to Python and open source software in general. One Pythonic example of
such a company is Digital Creations, which open sourced a lot of Python
code before it finally went all the way in open sourcing their Zope
framework. It would be hard to imagine something like this if Python was
covered by the GPL, though the GPL has its own famous examples (like Cygnus
for gcc).

License-pundit-for-5-minutes-ly yours,

Martijn




From xah at best.com  Mon Aug 23 10:06:09 1999
From: xah at best.com (Xah)
Date: Mon, 23 Aug 1999 07:06:09 -0700
Subject: Enough [Was Re: Why use Perl when we've got Python?!]
References: <1277478426-31364719@hypernet.com>
Message-ID: <37c15551$0$205@nntp1.ba.best.com>


----------
In article <1277478426-31364719 at hypernet.com>, "Gordon McMillan"
<gmcm at hypernet.com> wrote:


> John W. Stevens wrote:
>
> [ crap deleted ]
>
> Whatever your contributions to the python community, you're now in my
> killfile, along with a handful of pl hotheads who do the same to us.


Wow! Occasionally a righteous man stands-up from the rabble fearlessly
declare the obvious!!! I didn't know that there's such men in the Python
group too!

Sir, may I kiss your foot?

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 Department of philosophy, Bovine University



From I.Clarke at NOSPAM.strs.co.uk  Fri Aug 13 04:54:54 1999
From: I.Clarke at NOSPAM.strs.co.uk (Ian Clarke)
Date: Fri, 13 Aug 1999 09:54:54 +0100
Subject: help learning python source code
References: <7os22i$9s5$1@nnrp1.deja.com>
Message-ID: <37B3DD5E.4E9E3B1@NOSPAM.strs.co.uk>

Download and install the appropriate version of Python for your platform
(from http://www.python.org/) and look for the tutorial in the
documentation (it isn't hard to find!).  The tutorial will give you an
excellent introduction to Python and combined with the library reference
(also included) will tell you everything you need to know.  In order to
follow the tutorial you should probably have programmed in another
language before (even if only a simple language like Basic).

Ian.



From foo at bar.com  Wed Aug 11 16:48:24 1999
From: foo at bar.com (foo)
Date: Wed, 11 Aug 1999 13:48:24 -0700
Subject: Python tkinter with Tcl/Tk 8.1?
Message-ID: <7osg3n$l41$1@news.vanderbilt.edu>

I could not find any documentation on this in the FAQ or the tkinter pages,
so I am asking here. I have had to move to Tcl/Tk 8.1 for another software
package I am using on my computer. I have rebuilt Python 1.5.2 after getting
the new Tcl/Tk installed and (no great surprise) tkinter does not seem to
work (I can successfully import Tkinter from the python command line, but
attempting to run Tkinter._test() crashes python due to access violations
(reading wild pointers).

Is it possible to build the tkinter module successfully with Tcl/Tk 8.1?

Thanks,
Jonathan





From jstevens at basho.fc.hp.com  Fri Aug 13 22:53:42 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 19:53:42 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132027.OAA03925@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:27 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cc26@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :Which begs the question: Just how many of you Perl II (Perl 5.003, is
> :it?)
> :programmers are writing OO-Perl?
> 
> What the heck is "Perl II"?  

Perl 5 is not just Perl 4 with a few changes.  It's darn near a
new language, hence: Perl II (you know, like Modula 2, then Modula 3?).

> And what is "OO-Perl"?

Object Oriented Perl.

> How many of us write new OO libraries?
> Use existing OO libraries?  What?

Any or all of the above.

John S.




From m.faassen at vet.uu.nl  Sat Aug 28 14:01:25 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Aug 1999 18:01:25 GMT
Subject: A Plea For A Return To Sanity!!!
References: <000301bef12c$1bfd6940$ad2d153f@tim>
Message-ID: <7q985l$7bi$3@newshost.accu.uu.nl>

Tim Peters <tim_one at email.msn.com> wrote:
> [Bit Nik]
>> Another quick question, are there such things as PUGs? (Python User
>> Groups). I couldn't find any information on python.org...

> SIGs -- Special Interest Groups.  Check the "Community" and "PSA" links -- and
> sign up for the PSA while you're there.

No, no, SIGs something else again. He was asking about PIGs. I don't imagine
he was asking for groups discussing the intricacies of the class/type 
distinction just yet. :)

Regards,

Martijn




From mhammond at skippinet.com.au  Wed Aug  4 22:57:04 1999
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 5 Aug 1999 02:57:04 GMT
Subject: [Image-SIG] Proper application of the buffer interface
In-Reply-To: <14248.63286.73014.204886@weyr.cnri.reston.va.us>
References: <14248.63286.73014.204886@weyr.cnri.reston.va.us>
Message-ID: <006501bedeee$33ef5c20$1101a8c0@bobcat>

>   The context is this:  the t1python package (an interface to a Type1
> font renderer) has, in the past, been passing bitmaps back to Python
> as string objects.  In the next release, I may create a new type in
> the C layer that provides all the interesting stuff, and
> would like to
> be able to convert these "glyph" objects to PIL images and GTK+/GNOME
> compatible images.  Does it make the most sense for the glyph objects
> to offer the buffer interface to make more conversions possible
> without increasing the number of memory copies, or do I misunderstand
> the application of the interface?

I believe you are correct, although _all_ these packages need to support
the interface.  The good news is that due to tricks inside Python, they
already may.

As you mention, it is common to use Python string objects to move chunks of
binary data around.  The buffer interfaces now allow us to use _any_
object, and as long as it conforms to the buffer interface, you dont lose
anything by not using strings (and obviously gain whatever functionality
your object has)

The general idea is that PIL and other such frameworks can think in terms
of buffers.  Rather than PIL saying "give me a Python string with the raw
binary data", it can say "give me an object from which I can obtain a
buffer with the raw binary data".  Strings obviously still fit the bill.

The good news is that it is most common for extension modules to spell
"give me a string with the raw binary data" as "PyArg_ParseTyple("s#",
...);".  PyArg_ParseTuple has been upgraded to use the buffer interfaces,
and so have Python string objects.  Thus, whenever code uses
PyArg_ParseTuple in that way, they are already supporting the buffer
interface.

Thus, you could implement your new object, and define the buffer
interfaces.  This object could then be passed to any C extension function
that use PyArg_ParseTuple, and the extension module will still think it has
a "char *" pointer from an in-place string object.  In practice, this means
that automatically people will be able to say "file.write(your_object)"
etc. with your new object.

The problem remains, of course, for extensions that use the
PyString_Check(), PyString_AsString() etc functions.  If they were upgraded
to use the buffer interfaces, then the transition would be complete.

Just to extend my guesswork somewhat, there is a new built-in "buffer()"
function.  This returns a "buffer" object.  I speculate this should be used
in preference to Python strings when you have binary data.  As this buffer
object supports the buffer interfaces, they are basically as functional as
strings for this purpose, but clearly indicate the data is not really a
string!  This appears to be more a matter of style, and also paves the road
to Unicode - eg, it makes sense to convert any Python string object to
Unicode, but not necessarily a binary buffer.

>   I'd appreciate some input on this matter.  I hope to get the next
> release of t1python done before too much longer, and this is probably
> the biggest remaining question that I need to deal with.

Probably Greg and Guido are the only 2 with the real insight, as they
threashed out the details.  But Im pretty happy with my understanding (as
detailed above) of the issues.

Mark.




From philh at vision25.demon.co.uk  Wed Aug 11 17:46:41 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Wed, 11 Aug 99 21:46:41 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net>
Message-ID: <934408001snz@vision25.demon.co.uk>

In article <37B0E67D.27F8D2ED at prescod.net>
           paul at prescod.net "Paul Prescod" writes:
> Phil Hunt wrote:
> > In article <37B02965.B1FE0E8D at prescod.net>
> >            paul at prescod.net "Paul Prescod" writes:
> > > Have you looked into XUL,
> > 
> > Glade is a GUI application that allows you to build a window visually.
> 
> Okay, but I saw a reference somewhere to Glade having a declarative file
> format that could be converted into code to build the GUI. I was mostly
> talking about the declarative file format.

I think I will have to have a look at Glade. My understanding is that 
it uses its declarative file format to save/load UI definitions that
people are working on.

If Parrot could read/write Glade's format, that would be good, as
then people could write a UI in Glade and export it to any
backend that Parrot supports.

> > My system (provisional name: ``Parrot'') is a command-line application
> > that takes an ascii file as input, and outputs a file containing source
> > code in your favourite language, to build the GUI.
> 
> Glade does that too, as long as your "favorite language" is C. :)
> 
> I don't understand why you would take a code generation approach,
> though. My professional opinion is that code generation is yucky. :) It
> takes no more effort to write a "Parrot reader" for each language --
> especially if Parrot uses the language's built-in XML support (oops.
> scratch that). Seriously though, XML aside, why bother with code
> generation when you can just do everything dynamically? Read the parrot
> file and make the objects on the screen. The performance shouldn't be
> much different and the extra dynamicity will make your life easier and
> make the tool cooler to use. "I'll just generate a modified parrot file
> and reload."

That is a persuasive argument. One counterargument would be that 
any application that uses that approach would need a built-in
XML parser and Parrot run-time library, thus creating code bloat;
however, Parrot is explicitly intended to support multiple back-ends,
so there is no reason it couldn't support both approaches.

What I am hoping is that once I have written a few back ends
(Python/tkinter, and HTML will probably be the first, and the back
end API is somewhat stable, I will be able to interest other people
in contributing backends. The ultimate aim is a universal front end:
you learn it once, and you can then easily write GUI apps for
any programming language and toolkit. And if Parrot can read/write
files from Glade, or other graphical gUI builders, it will be the
best of all worlds.

> > >  * it can be trivially parsed with existing code written in several
> > > languages
> > 
> > True; I don't think a parser for Python will be too difficult to
> > write: my provisional spec only has 12 productions.
> 
> These specs always start out small. :)

Yep, fraid so :-)
 
> > True, but writing classes implementing recursive tree-structures
> > is almost trivial in Python. (One reason I like it).
> 
> Wait until you get to the TCL version!

I had a look at Tcl; I think I will pass on that one (and the Perl 
one) :-).

> > I will stick with my current format; but adding a feature to
> > import/export to XML shouldn't be too difficult -- then I'd have
> > the best of both worlds.
> 
> True enough. I think history shows that one syntax or another always
> ends up dominating, though. Lisp had two syntaxes once...

-- 
Phil Hunt....philh at vision25.demon.co.uk




From robin at jessikat.demon.co.uk  Sun Aug 22 21:18:20 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 23 Aug 1999 02:18:20 +0100
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com>
 <37BEC755.645A350D@fnmail.com>
Message-ID: <0Hg8yBAcFKw3EwjN@jessikat.demon.co.uk>

In article <37BEC755.645A350D at fnmail.com>, Per Kistler
<kistler at fnmail.com> writes
>Hi
>
>...but Tom has been attacked over and over again. I can really
>understand that he may loose temper after beeing treeted like
>a idiot again and again. And Tom is really not any dummy. 
>I suggest to treat the "bigger" guys in the computer community
>with more respect, which would automatically prevent any "discurse"
>like the preceeding one. If that's not possible, I suggest another
>way: Zen-meditation. This adds a exception handling wrapper
>around the language-syntax-torn mind:-)
>
>Cheers, Per.
>
>srenner at lycosmail.com wrote:
>
>> The reactions Mr. Christiansen evokes may have more to do with his style
>> than his content. It strikes me as pompous, pretentious, hostile,
>> autodidactic, preening, vain and unpleasant. Perhaps this is not the
>> right newsgroup for him.
>
>
A genius gets some allowances of course, but we've seen TC in other
contexts usually being (or seeming to be) arrogant, bigoted and plain
nasty. The Tcl and Python communities certainly know him well. As an
amateur of computer 'science' I have always been interested in computer
languages as well, but I don't think there's much I can tell the python
community about linguistics (computer or human). I was originally
appalled by the whole concept of whitespace meaning something (I hated
occam the crappy transputer language). Oh dear are there any transputer
fans left out there?

The L Walls & T Christiansens may operate in mysterious ways beyond our
ken, but genius doesn't need to bang drums.
-If dinosaurs had learnt perl they'd probably still be here-ly yrs- 
Robin Becker



From mal at lemburg.com  Fri Aug 27 04:16:07 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 27 Aug 1999 10:16:07 +0200
Subject: MS Access 97 class.
References: <7q4t2r$99j$1@nnrp03.primenet.com>
Message-ID: <37C64947.33FC3F94@lemburg.com>

Bryan Jernigan wrote:
> 
> Has anyone developed a class for manipulating MS Access 97 database tables?

Try one of ODBC, ADO, DAO, COM -- whichever suits you best.
Python supports them all (with more or less extra work).

The database topic guide has the pointers:
	http://www.python.org/topics/database

Or did I miss your point ?
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   127 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From stephan at pcrm.win.tue.nl  Tue Aug 17 04:06:25 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 17 Aug 1999 10:06:25 +0200
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com>
Message-ID: <m3wvuuu8zi.fsf@pcrm.win.tue.nl>

Ole-Kristian Villaboe <olevi at my-deja.com> writes:

> Is there a command to do this? (Can't find one but...)
> I've been writing an algorithm myself, but it's incredibly slow and I
> would like something quicker (of course :) )

This works for me:

import random

def randomize(l):
    length = len(l)
    for i in range(length):
        j = rand.randrange(i, length)
        l[i], l[j] = l[j], l[i]

Note that this randomizes the list in-place.

Greetings,

Stephan



From mlh at idt.ntnu.no  Thu Aug 12 10:12:16 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 12 Aug 1999 16:12:16 +0200
Subject: Why is CGI module so complicated?
References: <37AB04A8.C3403427@NOSPAM.strs.co.uk> <cWYr3.1329$EG4.226668@news.shore.net> <37B15BF8.B7204705@strs.co.uk> <mjhs3.1462$EG4.260559@news.shore.net>
Message-ID: <y0jiu6lf5q7.fsf@drue.idi.ntnu.no>

"Michael P. Reilly" <arcege at shore.net> writes:

> Ian Clarke <I.Clarke at strs.co.uk> wrote:
[...]

> 
> What would be _really_ good (IMO) is to have a "toDict" method that
> does this.

Why not just a factory function in the cgi module?

-----

import cgi

form = cgi.getFields() # Returns a dictionary version of FormStorage()

-----

To me, this would be intuitive... You wouldn't have to go through a
seemingly unnecessary class on the way.

> 
>   -Arcege
> 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From che at debian.org  Tue Aug 24 16:44:32 1999
From: che at debian.org (Ben Gertzfield)
Date: 24 Aug 1999 13:44:32 -0700
Subject: Python equivilent of pack()?
References: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com> <14274.61530.251401.762453@weyr.cnri.reston.va.us> <ytt6725sz5u.fsf@gilgamesh.cse.ucsc.edu> <14275.479.707949.774041@dolphin.mojam.com>
Message-ID: <ytt3dx8ucwf.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Skip" == Skip Montanaro <skip at mojam.com> writes:

    Skip> In fact, the Module Index is much more useful than the
    Skip> contents of the lib ref manual once you're fairly well
    Skip> settled in.  I used to generate an index myself before Fred
    Skip> started doing it (it was a pain to keep up with the TOC
    Skip> format changes!).  Now I simply link to

    Skip>     http://www.python.org/doc/lib/modindex.html

    Skip> from my private home page and skip the lib manual's front
    Skip> page entirely.  Most everything I need is no more than two
    Skip> or three clicks away.

You're absolutely right, Skip. The current Python lib ref manual front
page is not nearly as useful as that module index page.

Perhaps it's time for a reorganization? The main lib ref page has
gotten rather bloated, and it's pretty hard to find the necessary
information. Maybe we could go for a Yahoo-style categorization
rather than the huge delimited list we have now.

-- 
Brought to you by the letters C and T and the number 4.
"Moshimoshi. Kikoemasu ka?" "Kakenaoshimasu kara ne! 1-do kitte kudasai."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From mal at lemburg.com  Fri Aug 13 05:36:19 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 13 Aug 1999 11:36:19 +0200
Subject: portability question
References: <37B3B84F.F818F396@inka.de>
Message-ID: <37B3E713.88AA711@lemburg.com>

Friedrich Dominicus wrote:
> 
> About file and directory handling on different platforms.
> 
> I'm aware that I can read the source to find out. But maybe someone can
> give me some other hints about that.
> 
> How is it done in Python? Do you think the Python solution of doing it
> is ok?

All those operating system specific quirks are nicely hidden
by the standard os module. Path conversion is implemented in the
submodule os.path. Docs are available at www.python.org and
included in your Python distribution.

BTW, reading the source will give you an impression of how difficult
it sometimes is to get things right.... ah yes, and as TimP would
say: text files aren't portable. Open them in binary mode and
then do the line ends stuff in Python.

Hope that helps,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   140 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From dfan at harmonixmusic.com  Wed Aug 18 10:03:50 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 18 Aug 1999 10:03:50 -0400
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <14265.47748.241386.291144@buffalo.fnal.gov> <wkpv0mgna5.fsf@turangalila.harmonixmusic.com> <14265.58497.578663.958576@buffalo.fnal.gov>
Message-ID: <wkpv0l5gop.fsf@turangalila.harmonixmusic.com>

Charles G Waldman <cgw at fnal.gov> writes:

| Dan Schmidt writes:
|  > 
|  > Skip's method is fine if choice is allowed to equal i.  With that
|  > change, element exchange is no longer forced.  You can prove that
|  > it's a perfect shuffle by induction: the last element in the
|  > shuffled sequence has an equal chance of being any of the
|  > original elements, and so on.
| 
| Correct.  However, I'm curious, what is the definition of a "perfect
| shuffle"?  Just that each element of the permuted sequence has an
| equal chance of being any of the original elements?

Nope, you need to show that all permutations are equally probable.

| As I tried to point out in my earlier posting, there are plenty of
| ways of choosing permutations for which this condition holds, but
| still don't uniformly sample the symmetric group (set of all
| permutations).

You use induction.

We assume as the induction hypothesis that the method can randomize
n-1 elements perfectly.

Now, to randomize n elements, we pick one perfectly randomly to put in
the final spot, and randomize perfectly the other spots (by the
induction hypothesis).

There's a more formal discussion in Knuth, The Art of Computer Programming
vol. 2, sections 3.4.2 (algorithm P) and 3.3.2 (algorithm P).

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From mal at lemburg.com  Wed Aug 25 03:30:03 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 25 Aug 1999 09:30:03 +0200
Subject: [ANNOUNCE] M2Crypto 0.01
References: <37bd6fa4.0@news.cyberway.com.sg> <37BED351.55E7DA98@lemburg.com> <37bee546.0@news.cyberway.com.sg>
Message-ID: <37C39B7B.10BC9FF1@lemburg.com>

Ng Pheng Siong wrote:
> 
> According to M.-A. Lemburg <mal at lemburg.com>:
> > Ng Pheng Siong wrote:
> > > I am pleased to announce the release of M2Crypto 0.01;
> > > the package is available at http://www.post1.com/home/ngps
> >
> > Could you please tell me where that server is located and whether
> > it is legal to download the software from Germany ?
> 
> The server is located in Singapore. AFAIK, there is no law against
> export of cryptography from Singapore.

Great, then I'll take a look...
 
> > Anyway, mxCrypto development will continue to eventually include
> > most APIs of OpenSSL.
> 
> Hey, there are umpteen Java crypto libraries. Python can live with
> three or five. ;-)

Oh, I've got no problem with that :-) -- I just wanted to give
people already using mxCrypto a note about its status.

Cheers,
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   131 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From philh at vision25.demon.co.uk  Mon Aug 23 15:41:03 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 23 Aug 99 19:41:03 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <slrn7s1cug.1vk.irclark@latveria.castledoom.org>
Message-ID: <935437263snz@vision25.demon.co.uk>

In article <slrn7s1cug.1vk.irclark at latveria.castledoom.org>
           irclark at latveria.castledoom.org "Isaac" writes:
> On Sat, 21 Aug 99 23:56:16 GMT, Phil Hunt <philh at vision25.demon.co.uk> wrote:
> >It looks like version 0.0.1 of Parrot, my text-based GUI builder,
> >will be ready within a day or so. It doesn't actually do anything
> >useful yet; v0.0.1 is a technology demonstration version, which
> >is intended to verify that the basic design is more-or-less
> >sensible. Would anyone like to have a look at it, and possibly 
> >suggest some improvements? If so, tell me, and I'll put it on
> >my website.
> >
> >Some information about Parrot follows:
> 
> Very interesting indeed!  I was working on a similar program whose
> functionality would fit much of the description you give here.  My
> program reads in a description of a gui that's similar in concept
> to your par files, and outputs code that implements the gui, in
> either Motif, GTK, or the EZWGL

What's that?

> toolkit based on the backend chosen.   
> Currently the code is limited in the number of supported widgets. 
> I haven't tried to output anything but C code, but I don't see
> why other procedurally based languages wouldn't work.

My next step, once I'm happy with the parser and the HTML backend,
will be to write a Python/Tkinter backend. Then, I'll have written two
backends so I'll have a better idea of how to architect a backend 
framework -- I want to make backends easy to write, to encourage 
other people to contribute them.

> I haven't had time to work on my program lately, and the last time I
> looked at it, I really didn't like the syntax I'd chosen for the
> gui representation.   I'm using PCCTS to implement the parser for the
> language.

What language is your tool written in? AFAIK, PCCTS is a C parser
generator -- is this correct? How does it compare to yacc?

> I used a separate parser for implementing a language 
> for describing backends.  It's actually the language implementing part 
> that I find the most fun.  I barely learned enough about the various 
> toolkits to get the code output part working.

Yeah, the various toolkits can be quite hard to learn -- which is
why a gnerator is a useful idea.

> Looks like your seriously working on your program while I just can't
> find much time to continue.  (I just started an evening law school
> program.)  If you put the code up, I'd be happy to give you all the
> feedback I can. 

Looking forward to hearing it.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From stephan at pcrm.win.tue.nl  Tue Aug 31 11:33:47 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 31 Aug 1999 17:33:47 +0200
Subject: Newbie (read: pot. stupid) question on objects/namespaces
References: <8E339DC2FPaCmAnRDLM@news.dial.oleane.com> <m3iu5w5an5.fsf@pcrm.win.tue.nl> <8E33ADDF0PaCmAnRDLM@news.dial.oleane.com>
Message-ID: <m3emgk55ic.fsf@pcrm.win.tue.nl>

fredp at multimania.com.nospam (Fred Pacquier) writes:

> stephan at pcrm.win.tue.nl (Stephan Houben) said :

> >Note that in *both* cases you create a circular reference which means
> >that CPython cannot reclaim the class Bar or the instance of Baz.
> 
> Duh. Is this a total "no-no", or what ? If this is only for a handful of 
> instances that are supposed to persist through the entire program lifetime, 
> is it okay to explicitely del() them at the end or not ?

No, this was just a word of caution!
If it's just a handful of instances, don't bother.

However, if you want to get rid of them, del'ling them doesn't work.
You must break the circularity to get them garbage collected.
In my original example:

> >class Baz:
> >    def __init__(self):
> >        self.x = MyClass(self)

,when you have created an instance:

baz = Baz()

,you can break the circularity by either doing:

baz.x = None

or

baz.x.owner = None

But don't do this in the __del__() method of Baz!
That's not good enough. You have to do this in order to
get __del__() called in the first place.

Now, if you use JPython, you have real garbage collection, and
all this is a non-issue.

Greetings,

Stephan



From bizzaro at bc.edu  Tue Aug 10 23:00:16 1999
From: bizzaro at bc.edu (J.W. Bizzaro)
Date: Wed, 11 Aug 1999 03:00:16 +0000
Subject: ANNOUNCE: pygtk-0.6.2 and gnome-python-1.0.4
Message-ID: <37B0E740.2A6667D8@bc.edu>

      ANNOUNCE: pygtk-0.6.2 and gnome-python-1.0.4

James Henstridge is the author of the Python language bindings to the GTK+ and
GNOME GUI widget sets.  His most recent release includes quite a few extras. 
His message follows...

I have finally got round to releasing pygtk and gnome-python.  You
probably already know what is new, but if not, here are the major points:
  - fixed threading support (thanks to Matt Wilson and Paul Fisher).
  - A pygtk.h header file to make it easier to write wrappers for other
    add on widgets.
  - Wrappers for GtkGLArea and libglade.  To actually make use of the
    GtkGLArea widget, you will need an OpenGL binding for python, such as
    PyOpenGL.  There are some examples of its use in examples/gl (there is
    a port of the Mesa gears example in that directory, among others).
  - NumPy support in GdkImlib.  You can create an image object from a
    MxNx3 numpy array, and get numpy array representing the data in an
    image object, which you can manipulate.  This code is based on stuff
    sent in by Travis Oliphant.
  - Support for gdkrgb, so you can draw rgb and greyscale data to a
    drawable.  I have also added a numpy array variant called draw_array.
    It accepts unsigned byte arrays of the following forms:
      + MxN or MxNx1: treated as greyscale data.
      + MxNx3 or MxNx4: treated as RGB data.
    The numpy wrapper will also work with some non contiguous arrays, such
    as arr[a:b, c:d, :] (sliced along first two axes).
  - methods for the GdkWindow object for setting and reading properties.
    Also, GdkWindows now have an xid attribute which can be used to get
    the X window ID of the window.
  - You can now pass extra parameters to idle and timeout functions, like
    you can for signal handlers.  There is a gtk_signal_connect variant in
    pygtk that is similar to gtk_signal_connect_object as well now.

Gnome-python mainly contains bug fixes rather than new features.

You can get the new release at
  ftp://ftp.python.org/pub.contrib/Graphics/       (should be there soon)
  ftp://ftp.gnome.org/pub/GNOME/sources/latest/       (gnome-python only)
  ftp://ftp.gtk.org/pub/gtk/python/      (pygtk only -- not uploaded yet)
  ftp://ftp.daa.com.au/pub/james/python/

Remember to report any bugs you find.

James.
--
Email: james at daa.com.au
WWW:   http://www.daa.com.au/~james/
--

More information on Python bindings to GNOME-related toolsets can be found at
the PyG Tools Information Page:

    http://theopenlab.uml.edu/pygtools/



From jkmuir at uniserve.com  Sun Aug 22 13:34:18 1999
From: jkmuir at uniserve.com (Justin Muir)
Date: Sun, 22 Aug 1999 12:34:18 -0500
Subject: Tkinter and classes
Message-ID: <37C0349A.EF5EEA6D@uniserve.com>

Hi all.   I'm just fiddling about with Tkinter buttons and labels trying
to get the text in a Label to change when I press a button, this is
essentially what I've done:


class Helloworld:
    def __init__(self, master):
            _lblstr = StringVar()
            _lblstr.set("hello world")

            frm = Frame(master)
            frm.pack()

            but = Button(frm, text='Hello!', command=setLabel)
            but.grid(row=0)

            lbl = Label(frm, textvariable=_lblstr)
            lbl.grid(row=1)

    def setLabel(self):
            _lblstr.set("HELLO WORLD!!")

This will run, but barfs out on the callback.
Am I misunderstanding the relationship between the callback to the
namespace of the class?

Thanks!

Justin




From skaller at maxtal.com.au  Sun Aug 15 12:15:46 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sun, 15 Aug 1999 16:15:46 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl>
Message-ID: <37b6dea0.342930367@news.triode.net.au>

On 11 Aug 1999 08:21:14 GMT, m.faassen at vet.uu.nl (Martijn Faassen) wrote:

>It sounds like my (vaporware) Swallow idea. The difference with Viper
>is that Viper apparently is going to do type inference, while Swallow
>would (at least in the first stages) make do with nonintrusive type
>declarations, somewhat like this:
>
>__types__ = { 
>  foo : functionType(args={a: intType, b: intType}, returns=listType(intType)),
>  bar : functionType(args={}, returns=None)
>}

	This is neatly compatible with existing Python code.

>> The interpreter aims to be compatible with 'pure python':
>> python not using external C modules, and not fiddling
>> implementation details too much. Some of the fiddles
>> will work, and some will not. 
>
>This is a good idea, as pure Python will often still be the development
>language (while Swallow/Viper would only be invoked to compile and
>optimize things later on, at least in the early stages).

	Originally, I planned to recommend development
in Python, and just using Viper for compilation: that is,
I intended to ignore any sophisticated debugging support.
However, I'm beginning to swing the other way now, and
try to provide _better_ debugging than Python.

	The reason is pragmatic: without the debugging
support, I can't debug Viper itself. 

>> Compatibility:
>> 	4) range is a keyword
>
>Why this? Perhaps instead you could make 'range' and some other builtin
>functions 'special functions'. 

	Yes, that's the idea.

>The idea is that like module variables
>(and some other set of things perhaps) they cannot be rebinded to something
>else, so that the optimizer can do its stuff.

	That is what I meant by saying it was a keyword.

>> Extensions:

>I wouldn't aim for *any* extensions to start with. In order for such a
>project to succeed you want Viper code to run in Python and vice versa, 
>unless there's serious messing with internals. At least make it possible
>to turn these extensions *off* so they'll be errors.

	There are two kinds of extensions: those requiring
extra keywords, and those that 'fit naturally' into python.

	Extensions requiring extra keywords can be turned
off with a compiler switch that disables the lookup in the lexer
that maps names to keywords. Then the parser will not
recognize these extensions.

	The other kind of extensions include things
like lexical scoping and slicings, where existing or extended syntax is given new
semantics. Sometimes (hopefully rarely) this may lead to an
incompatibility. For example nested functions in Viper
are bound to the activation record of the enclosing
function. As such, variables in the enclosing scope will
hide globals, changing the semantics.

>Hm, I wish you luck, especially with the type inference...

	Preliminary tests indicate reasonable results
determining types. In critical cases such as inner loops,
it is usually possible:

	s = ""
	for  i in range(1,10): s = s + str(i)

Obviously i is an integer here, and s a string.

>Does viperc compile Python to C code? 

	I don't know yet. In the first instance,
a run time is still needed that works like existing Python,
where type inference fails. Therefore, I still need an
interpreter. So Viperi, the interpreter, is the first step.
[It doesn't support 'global' at the moment, so I cannot
yet run PyStone. I'd like to eliminate 'global']

	I will try to optimise it first, before considering
how to build the compiler, and what back ends to implement.
It is possible the first back end will actually general Ocaml
rather than C (since Viper is currently implemented in ocaml).

>> 	Generally, compiler optimisations
>> will go hand in hand with restrictions -- to optimise
>> something, information is required, and that will
>> require giving up some dynamism.
>
>So you could do with Swallow's "type system" (big word for nothing there),
>right? 

	Yes. It is useful to have optional typing, to 
help make the type inference work. I was considering:

	def f(x:int): ...

as the syntax ( because this is fairly conventional,
and a pure extension). However, it isn't
backward compatible, whereas your idea is,
so I will look at it. [The method resembles the
'traits' templates in C++ STL]

>> 	It is difficult to optimise individual
>> modules. 

>Hm, but the possibility to optimise individual modules is what attracts
>many people to this idea. 

	I know.  However, existing attempts to
build a python compiler have not succeeded in
producing code that is much more efficient than
the bytecode interpreter. 

	Why is this? I suspect it is because
a function in Python is often unintentionally
more polymorphic than actually desired: for example:

	def add(i,j): return i + j

Due to the unfortunate overloading of + on
numbers and strings, it is not possible to
determine if 'add' adds numbers or concatenates
strings. However, this can be determined by examining
_calls_ to the function add.

It is possible that both are done: the function is actually
used polymorphically .. but not necessarily in the
same program

	Perhaps a _worse_ case is this:

	def get_attribute(a,b):
		try: return get_attr(a,b)
		except: return None

Here, the return type could be anything at all.
If we knew 'a' was an object with only string
attributes, the return type would be either
string or None.

	In that case, the function is monomorphic,
but the return type is the sum of string and NoneType.
	
>  * Swallow now optimises these functions by translating them to C, and
>    you can use them in Python (just like C extension modules).
>
>For classes you'd need some extra support (and more restrictions) but that
>was phase 2. My advice is to start as small and unambitious as possible
>for such an ambitious project. :)

	I agree. However, it turns out adding extensions is often
trivial, whereas providing compatibility is often difficult. Similarly,
it may be more difficult to generate a function than a whole
module, and more difficult to do that than build a whole
program.

	In particular, building a whole program is easiest,
since it doesn't require interfacing to python. That interface
is non-trivial: binding to the C API, and then the bytecode
interpreter, and then again, the result is different if you
were using JPython.

	So it may be easier to just provide
a binding at a single level, namely 
'pure python' source code, than try to build something
that interfaces to existing Python.

	In addition, Python 1.6 may have different
internals, and so may Python 2.x. This creates
a maintenance nightmare which may make the project
less useful.

	OTOH, many python 'built in C modules'
have corresponding Pure Python sources, for example
the string module. In that case, if the compiler works,
there is no need for the C module.

	Another example is Lemburgs mxTools:
I have a 'Pure python' implementation of it,
so I don't need the C version, thus eliminating
a  maintenance problem.

	Another example (which is now working):
Python has a module 'exceptions'. I have no idea
if Python actually compiles this module and uses it,
however, Viperi currently does. That is, it actually
imports the exceptions module and uses the
classes in it for exceptions.

	If Guido adds a new exception,
or changes the semantics, implementation,
or interface, I have a rock solid python compatible
specification to implement, and half the work
will be done by Viper itself. [Not all, since the
internal exceptions stll have to be mapped to the
python ones]

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From bwarsaw at cnri.reston.va.us  Mon Aug 16 14:10:27 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 16 Aug 1999 14:10:27 -0400 (EDT)
Subject: [REPOST] Wash. D.C. Area: DCPIGgies
Message-ID: <14264.21523.767752.968874@anthem.cnri.reston.va.us>

Notice to all Washington DC area Python users:

We are pleased to be resurrecting the DC area Python interest group
meetings, now called DCPIGgies.  We will be having our first meeting
at CNRI in Reston, Virginia on Tuesday August 31, 1999 from 7:30pm to
9:30pm.  The aim is to keep this meeting pretty technical.

To that goal, we're going to have two exciting presentations.  For the
first hour, the Digital Creations guys will be giving a talk on Zope
2.0's unique object model (acquisition, run-time classes,
transactional objects, etc.) including how Python programmers might
use this in their own non-Zope projects.

In the second hour, Andrew Kuchling will be talking about something
XML-ish, probably processing the Open Directory Project's
(www.dmoz.org) RDF dumps.

There will be question and answer periods after each talk, and we may
have an "open phones" going later than 9:30 if there's interest and if
Guido or I am still awake. :)

Pizza, salad, and soda will be provided.  We are asking for a
voluntary $5 donation at the door to cover the costs of food.  Please
RSVP by email to bwarsaw at cnri.reston.va.us.  Let me know how many plan 
to attend and whether you will be eating pizza.

Directions to CNRI can be found on the Web:

    http://www.cnri.reston.va.us/directions.html

Feel free to repost this announcement.

See you then!
-Barry



From mlh at idt.ntnu.no  Tue Aug 24 07:55:51 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 24 Aug 1999 13:55:51 +0200
Subject: Unsorting(randomizing) a sequence
References: <000501beedf3$eb737200$902d2399@tim>
Message-ID: <y0j3dx9z92w.fsf@vier.idi.ntnu.no>

"Tim Peters" <tim_one at email.msn.com> writes:

> [Aahz Maruch]
> > ...
> > Here's another question: I know the *period* of whrandom() is very
> > long, but what is the granularity/precision of it?
> 
> It combines 3 sequences each of which have no better than 15-bit resolution.
> Analyzing the combined effect is tricky.  You need to define your requirements
> precisely, then write a realistic simulation independent of your app to test
> whether your requirements are met.  Every non-trivial application of
> pseudo-random numbers should do that.  Collecting random <wink> info about the
> period and resolution is no substitute; e.g., the lattice structure of
> random.random may kill you no matter how long its period or fine its
> resolution -- or it may be irrelevant.  It depends on every detail of your app.

Hm. AFAIK, some OSes (like Linux) have random methods that use such
things as the internal clock and samplings of user keystroke timings
etc... Does Python utilize these when available? These should be
closer to truly random than pseudo-random - no?

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From mbf2y at my-deja.com  Mon Aug 16 12:29:57 1999
From: mbf2y at my-deja.com (mbf2y at my-deja.com)
Date: Mon, 16 Aug 1999 16:29:57 GMT
Subject: Difficulty configuring web server for Python
References: <7p9855$5tg$1@nnrp1.deja.com> <19990816171506.K21639@xs4all.nl>
Message-ID: <7p9e9l$av1$1@nnrp1.deja.com>

Argh!

When I read your message, I was saddened because everything you listed
was stuff I'd already tried... I had already found the ^M lurking in the
top line of my python script and deleted it, and I'd already assumed
that the webserver was chroot-ed and had tried changing my first line to
a different path...

problem was, I did them in the wrong order.  I tried the corrected path
to python and that didn't work, so I undid that, and then I deleted the
ctrl-m and that didn't work.

After reading your post, I went to the perl script that I had downloaded
and changed it's path to the same full-path I was using and all of a
sudden the perl script bombed.  So I realized what had happened (I'd
tried the fixes in the wrong order) and voila, I get my hello world.

Much much thanks for the quick replies I got to my query.  I really
didn't want to have to learn that other scripting language ;-).

-Fred

---
Michael "Fred" Fredericks
PhD Student, Computer Science
University of Maryland, College Park
fred.at.cs.dot.umd.dot.edu ---- I NEVER check my Dejanews mail


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From skip at mojam.com  Mon Aug 16 16:49:24 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 16 Aug 1999 15:49:24 -0500 (CDT)
Subject: Python CGI Scripts
In-Reply-To: <Pine.LNX.4.04.9908162019270.3026-100000@natal.stamp-coin.com>
References: <37B7C061.3824206C@objectzoo.com>
	<Pine.LNX.4.04.9908162019270.3026-100000@natal.stamp-coin.com>
Message-ID: <14264.31019.782265.1744@dolphin.mojam.com>

    Jonathon> 	A Python equivelent to the Perl CGI.pm module would be a
    Jonathon> 	very usefull thing to have/do.

What does Python's existing cgi.py module not do that you would like it to?

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098




From andrew at starmedia.net  Thu Aug 19 16:31:02 1999
From: andrew at starmedia.net (Andrew Csillag)
Date: Thu, 19 Aug 1999 20:31:02 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
		<37BC1ED4.A29883BC@appliedbiometrics.com> <37bc2588@cs.colorado.edu>
Message-ID: <37BC6986.A66B044E@starmedia.net>

Tom Christiansen wrote:
>
> As for the tuple thing, why can I say
> 
>     print "You have $%0.2f bucks" % (4.5/100)
> 
> How is that a singleton tuple? 
It's not.  For format strings, if the item (here 4.5/100) is a single
thing, the items need not be a tuple. i.e.

   "You have $%0.2f bucks" % 234.234

but if you want more than one item to be formatted, you have to use a
tuple

   "you're name is %s and you have $%0.2f" % ("Tom", 234.234)

> Don't I have to use
> a trailing comma?  It won't let me!  Why is
> 
>     print "You have $%0.2f bucks" % (4.5/100,)
> 
> illegal, 
No, it's perfectly fine.  Didn't try it, I guess.

>>> print "You have $%0.2f bucks" % (4.5/100,)
You have $0.04 bucks
>>> 


but
> 
>     print "You have $%0.2f bucks" % ((4.5/100,) * 2)
This, on the other hand is wrong since ((4.5/100,) * 2) yields (0.045,
0.045) and you only have one formatting bit in the string (only 1
%0.2f).  If you had 
print "The New York Lotto jackpot is now $%0.2f bucks... $%0.2f" %
((4.5/100,) * 2)
That totally works.

> 
> mandatory?  Was this (elt,) thing just an, er, unforeseen design
> misfeature? Yes, I tripped me up again. :-(
Try reading the docs...
-- 
"Programmers are just machines that convert coffee to executable code"
   -Unknown

"Drew Csillag" <drew_csillag at geocities.com>



From earlybird at mop.no  Sun Aug 22 04:03:59 1999
From: earlybird at mop.no (Alexander Staubo)
Date: Sun, 22 Aug 1999 10:03:59 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <37bc313b@cs.colorado.edu> <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us> <019301beeaea$004c4d80$f29b12c2@secret.pythonware.com> <37bd4876@cs.colorado.edu>
Message-ID: <MPG.1229cd5eed750a81989687@news.mop.no>

In article <37bd4876 at cs.colorado.edu>, tchrist at mox.perl.com says...
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.python, 
>     "Fredrik Lundh" <fredrik at pythonware.com> writes:
> :as an aside, it's a bit sad that Tom didn't read the book's
> :preface, where Mark says, among other things:
> 
> It's a shame that so many of you should be so unbelievably harsh on me
> when *I'm* the one who with malice toward none and full of revulsion
> at the flames causes by *YOUR* community's ignorant attacks on mine,
> that *I* made the honest effort to learn *YOUR* language, and that I
> furthermore for your benefit not my own then took the additonal trouble
> to write-up my experiences and observations gleaned

With all due respect -- not being an active part of this discussion, I 
have observed from afar no flames or unreasonable critique, bar an 
obviously benign "Try reading the docs". Do I sense a whiff of paranoia 
in the air?

Your comments about manpages seem to imply that "man" is the defacto 
standard for documentation; it simply isn't. The quasiphilosophical idea 
that "if it's not in the manpages and not in the Lutz book, it doesn't 
exist" [my phrasing] sounds a bit like the tantrums of a person too 
comfortable in his own land to accept the value of others; "If it 
doesn't have McDonald's, it's not real country."

Your tool isn't everybody else's tool. Personally I don't even know what 
"apropos" is. Even if I made an effort at using it, it doubt it would 
make _my_ workday any easier; I have my own tools, and if I had to use 
man (which isn't part of my toolset), I would nonetheless submit to its 
archaic interface and use it, just as you ought to submit to the 
inflexibilities of the HTML documentation. No need to invoke a fatwah on 
anybody; we live to learn about and deal with our differences.

>while five hundred
> miles away from the nearest tree or internet link.

Python -- or the binary distribution at any rate -- comes with adequate 
HTML documentation in the box, no tree or Internet link required.
 
[one pompous declamation of self-importance, two generalizations, three 
incriminations against the community, one incident of improper language, 
and three cynicisms generously deleted]

> 
> --tom
> 

-- 
Alexander Staubo             http://www.mop.no/~alex/
"QED?" said Russell.
"It's Latin," said Morgan. "It means, 'So there you bastard'."
--Robert Rankin, _Nostramadus Ate My Hamster_



From arcege at shore.net  Sun Aug 15 11:29:56 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Sun, 15 Aug 1999 15:29:56 GMT
Subject: Enough [Was Re: Why use Perl when we've got Python?!]
References: <1277478426-31364719@hypernet.com> <37b58c4a@cs.colorado.edu>
Message-ID: <U1Bt3.143$mV.53412@news.shore.net>

Tom Christiansen <tchrist at mox.perl.com> wrote:
:      [courtesy cc of this posting mailed to cited author]

: In comp.lang.python, gmcm at hypernet.com writes:
: :John W. Stevens wrote:
: :[ crap deleted ]
: :
: :Whatever your contributions to the python community, you're now in my 
: :killfile, along with a handful of pl hotheads who do the same to us.

: That man, along with Ian Clarke, have stirred up the biggest flame war
: in the Perl newsgroup that we're seen for a long time, inflicting their
: standard overrighteous prattle on us.  I realize that they aren't truly
: representative of Python people in general: several others have been
: entirely reasonable and educative.  I'm not sure hoter users realize
: this, however.

There was an attempt on c.l.p to divert the issue from c.l.p.m; much to
our collective regret, it didn't work (obviously).

  -Arcege




From bennettt at am.appstate.edu  Mon Aug 23 11:10:33 1999
From: bennettt at am.appstate.edu (TMGB)
Date: Mon, 23 Aug 1999 11:10:33 -0400
Subject: Python / Cool looking applications?
References: <m3ogfz9mx6.fsf@intergate.bc.ca>
Message-ID: <37C16469.687EDE58@am.appstate.edu>

In addition to Mr. Clarke's suggestion, seeing that cool-looking is in
the eye of the beholder I'll just suggest an implementation you may want
to look at.  If you are using Linux 5.2 (I'm not sure about other
versions) you may want to look at the netconfig setup program.  If you
run it and then run ps from a term window you should see "python 
path2file"  where path2file is the path and file being loaded in python
to run the netconfig program.  If it is a .py file you should be able to
bring it up in an editor to view it.

Thomas

Gerald Gutierrez wrote:
> 
> Hi all.
> 
> I've been looking into using Python as UI glue and at Python's imaging
> extensions. Are there any examples of very cool looking interfaces or
> otherwise well-designed interfaces that are built in Python?
> 
> Also, does anyone know of any applications that use Python's imaging
> extensions?
> 
> Thanks very much.
> 
> Please forward replies to gutier at intergate.bc.ca.

-- 
----------------------------------------------------------------------
Thomas McMillan Grant Bennett           Appalachian State University
Computer Consultant II                  University Library
bennettt at am.appstate.edu               
http://www.library.appstate.edu/admin/
Voice:  828 262 6587			FAX:    828 262 3001

Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
operating system that was originally coded for a 4-bit microprocessor.
 - Chris Dunphy     Boot Magazine



From aaron_watters at my-deja.com  Sun Aug  8 00:52:44 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Sun, 08 Aug 1999 04:52:44 GMT
Subject: ANN: Gadfly bugfix test release
References: <7ns85r$cla$1@nnrp1.deja.com>
Message-ID: <7oj2ep$2p2$1@nnrp1.deja.com>

In article <7ns85r$cla$1 at nnrp1.deja.com>,
  aaron_watters at my-deja.com wrote:
> Gadfly users and others: Please try one of the following
> files
>
>    http://www.chordate.com/kwParsing/kwPtest.tgz (tar/gz version)
>
> or
>
>    http://www.chordate.com/kwParsing/kwPtest.zip (zip version)
>
> I believe this replacement fixes the following longstanding bugs:

I just uploaded these *again*.  One new bug is fixed
[recovery problem that is impossible, that someone ran
into anyway] and an old bug that wasn't actually fixed
before [unique indices should correctly prevent non-unique
inserts] is now fixed, I think.

Please try it out.  I'll make it the official release
after it's stable.  Thanks for previous feedback!!

Two more comments:
  1) I just decided to go to the O'Reilly Open Source
     Conference: does anyone want a hotel roommate?

  2) In my efforts for gadfly 2 I came up with the following
     benchmark

def test():
    range1 = range(100, 30000)
    print "testing loop"
    from time import time
    now = time()
    result = range1[:]
    for i in range(len(range1)):
        result[i] = (range1[i], 'x')
    elapsed = time()-now
    print "elapsed", elapsed
    print;
    print "testing map"
    now = time()
    result = map(None, range1, ['x']*len(range1))
    elapsed = time()-now
    print "elapsed", elapsed

def test2():
    range1 = range(100, 30000)
    print "testing loop"
    from time import time
    now = time()
    result = range1[:]
    for i in range(len(range1)):
        result[i] = range1[i]-3
    elapsed = time()-now
    print "elapsed", elapsed
    print;
    print "testing map"
    now = time()
    from operator import sub
    result = map(sub, range1, [3]*len(range1))
    elapsed = time()-now
    print "elapsed", elapsed

test()
print
test2()

On my underpowered win95 machine this prints
testing loop
elapsed 1.10000002384

testing map
elapsed 0.330000042915

testing loop
elapsed 0.439999938011

testing map
elapsed 0.330000042915

Does this mean that map is always better than for
loops even when it looks stupid?  Inquiring minds
want to know.
   -- Aaron Watters

===
"Winston, you are drunk."
"And you madame are ugly, but I shall be sober in the morning."


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From training at eiffel.com  Wed Aug 11 16:22:33 1999
From: training at eiffel.com (Interactive Software Engineering, Inc.)
Date: Wed, 11 Aug 1999 13:22:33 -0700
Subject: Design by Contract Seminar Series
Message-ID: <7osm11$b2i$1@news.rain.org>

ISE's EXCLUSIVE SEMINARS ON DESIGN BY CONTRACT AND THE COMPONENT REVOLUTION.
A ONE-DAY COURSE BY BERTRAND MEYER.

If you are involved in software development, either as a manager or as a
developer, you know how crucial it is to improve software quality and
productivity. In spite of the major advances of recent years, especially
Object Technology, it is still much too hard to produce correct, reliable
systems at a reasonable cost.

This state of the art seminar is a unique opportunity to hear about the
approach that is revolutionizing the software industry - not from an outside
observer, but from one of the key originators of the discipline, Dr.
Bertrand Meyer, one of a handful of object- oriented movers and shakers
whose work made modern object-oriented technology possible.

"Design by Contract" is the major technological breaktrough in software
engineering in the past 15 years. This course introduces Design by Contract
and shows why it can, like no other technique since the invention of classes
and objects, dramatically affect the way we develop software, and why it is
a required condition for achieving the promise of the Component Revolution.

For more information about the seminar, as well as locations, please visit
our Web site at http://www.designbycontract.com. You can also contact us by
email: training at eiffel.com.

Interactive Software Engineering, Inc
ISE Building
270 Storke Road, 2nd Floor
Goleta, CA 93117, USA
t. 805 685-1006
f. 805 685-6869






From aahz at netcom.com  Mon Aug 16 12:07:59 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Aug 1999 16:07:59 GMT
Subject: C's isprint() concept?
References: <000201bee78f$6bbf6ec0$f22d2399@tim>
Message-ID: <7p9d0v$m9c@dfw-ixnews11.ix.netcom.com>

In article <000201bee78f$6bbf6ec0$f22d2399 at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>[Aahz Maruch]
>>
>> Oh, sure.  What isn't clear from the way you write this is whether
>> a run of multiple non-printable characters should be replace with
>> a single space.  Here's a regex that I created recently as part of
>> some code to detect binary documents:
>>
>>    # Everything that isn't CR/LF or 7-bit normal characters
>>    reBinary = re.compile ( r'[^\r\n\x20-\x7F]' )
>
>Take a look at string.translate.

Ah.  Interesting.  I'm using re.sub because we're still on 1.5.1, but
we're supposed to move to 1.5.2 Real Soon Now.  How does
string.translate compare with the speed of re.findall?  (I'm only
interested in the number of matches; I don't intend to *do* anything
with the matches.)
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From martinp at mincom.com  Mon Aug  2 21:12:46 1999
From: martinp at mincom.com (Martin Pool)
Date: Tue, 03 Aug 1999 11:12:46 +1000
Subject: Python and Red Hat Linux 6
References: <379CD547.22A10A50@postoffice.pacbell.net> <379CD5C2.5BA096A2@mail.danbbs.dk> <7nirk7$icq$1@oak.prod.itd.earthlink.net> <37A5F220.B83EEF99@boi.hp.com>
Message-ID: <37A6420E.B0A24A24@mincom.com>

Bill Anderson wrote:
> Spud wrote:
> >
> > Yeah, using #!/bin/env/python should help, but Red Hat should support
> > pound-bang (#!) anyway. You might want to check in with tech support.
> >
> 
> Uhmm ... it is not RedHat, but a shell thing, and if #! didn't work,
> neither would #!/bin/env/python.

I think you mean

  #!/bin/env python

The command you're runing is /bin/env, and you're passing it the
parameter "python".  This doesn't work as well as it might because some
systems put env(1) in /usr/bin or /etc (!).  

Some systems are known to want a space after the bang so that they can
match the first four bytes as a magic number, but I don't think this
applies to GNU/Linux:

  #! /usr/bin/python

On GNU/Linux, the FSSTND ought to specify this: I think /usr/bin/python
should be safe because Python is standard on most distributions these
days.

-- 
 /\\\  Mincom | Martin Pool          | martinp at mincom.com
// \\\        | Software Engineer    | Phone: +61 7 3303-3333
\\ ///        | Mincom Ltd.          | 
 \///         | Teneriffe, Brisbane  | Speaking for myself only



From jeremy at cnri.reston.va.us  Tue Aug 17 12:14:55 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Tue, 17 Aug 1999 12:14:55 -0400 (EDT)
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <14265.29899.470021.309355@dolphin.mojam.com>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
Message-ID: <14265.35455.590852.517197@bitdiddle.cnri.reston.va.us>

I've rather fond of the following approach:

import random

def randomize(a, b):
    return random.choice([-1, 0, 1])

l = range(2, 20)
l.sort(randomize)

print l

[2, 3, 4, 5, 19, 6, 17, 13, 16, 7, 8, 18, 10, 14, 9, 12, 11, 15]

No idea how fast this is, though not very I expect.

Jeremy



From fredrik at pythonware.com  Tue Aug 24 04:39:31 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 24 Aug 1999 08:39:31 GMT
Subject: gcc optimization breaks NumPy?
References: <7psar7$kt1$1@nnrp1.deja.com>
Message-ID: <009f01beee0c$30b48df0$f29b12c2@secret.pythonware.com>

John Fisher <jfisher at are.berkeley.edu> wrote:
> I was getting consistent but inexplicable segmentation faults with the
> following code, using NumPy's C API to move data between Python and C
> (for sparse matrix multiplication; the data structure is from Meschach,
> in case that's relevant).  I found the problem to be caused by gcc's
> optimization options.

well, there's also a little strange thing in that?
code of yours:

> PyObject * MakeFromMes(SPMAT *in) {
>   int i, j, m, n, nzs, count, pos, therow, thecol;
>   int dummy[0];

(I didn't even know you could have arrays
with zero elements...)

>   double elem;
>   PyArrayObject *pr, *ir, *jc;
>
>   m = in->m; n = in->n; nzs = 0;
>
>   /* Count the number of nonzero elements of in
>      ISZERO is a macro to test for "good enough" floating point 0 */
>   for(i = 0; i < m; i++) {
>     for(j = 0; j < n; j++) {
>       if(!ISZERO(sp_get_val(in, i, j))) nzs++;
>     }
>   }
>
>   dummy[0] = nzs;

but here you assume it has one element more than
you requested...

depending on the stack layout, this may smash one
of the PyArrayObject pointers.  looks like a great
way to get a segmentation fault ;-)

</F>




From rgetter at usa.net  Sat Aug 28 01:02:31 1999
From: rgetter at usa.net (Robert Getter)
Date: Sat, 28 Aug 1999 01:02:31 -0400
Subject: MS Access 97 class.
References: <7q4t2r$99j$1@nnrp03.primenet.com>
Message-ID: <MPG.12313777d073f06989685@news-server>

In article <7q4t2r$99j$1 at nnrp03.primenet.com>, yerny at primenet.com says...
> Has anyone developed a class for manipulating MS Access 97 database tables?
> 
> Why reinvent the wheel?
> Thanks,
> Bryan
> 
> 
> 
I've done some really cool things manipulating unix apps with Python, 
TNVT, Access 97 and COM. You can do just about anything in Python that 
you could have done from within Access. Download the Access Snapshot 
Viewer and you can save reports to a file.

Install win32all, then:

from win32com.client import Dispatch

app = Dispatch('Access.Application')
 -- or --
dbe = Dispatch('dao.dbengine.35')

then look at the object reference in the Access on line help and go to 
town. I haven't been able to get Python to respond to COM events, but for 
Access, that isn't a problem.

Rob Getter
rgetter at usa.net



From keithmur at mindspring.com  Thu Aug 19 15:04:24 1999
From: keithmur at mindspring.com (Keith G. Murphy)
Date: Thu, 19 Aug 1999 14:04:24 -0500
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net>
Message-ID: <37BC5538.CEAFFA0F@mindspring.com>

Ben Caradoc-Davies wrote:
> 
> [strange programming]
> The most confidence uninspiring piece of code I ever had the displeasure to
> maintain contained something like this (allegedly C++, but mainly it's
> intersection with C):
> 
>         i = 0;
>         while( i < max_index ) {
>                 /* some copying or update (not modifying i) goes here */
>                 i = i + 1;
>         }
> 
> This was written by a *very* expensive consultant working for a large
> multinational accountancy firm.
> 
I think I can explain this.  The original code was written to compile
not under C++, but under a little known and short-lived language known
as "C=C+1".  It was to C++ as Python is to Perl.



From beall at psych.ucsb.edu  Tue Aug  3 20:12:41 1999
From: beall at psych.ucsb.edu (Andy Beall)
Date: Tue, 03 Aug 1999 17:12:41 -0700
Subject: building embedded classes as in C/C++
Message-ID: <37A78579.9856D16@psych.ucsb.edu>

Hi,

I'd like to build an embedded structure (Python class) much like I do in
C if possible.  When I do the following, multiple instances of my final
class overwrite each others sub-class values!

-------------------------------------------------------------------------
class Point:
	x = 0
	y = 0
	z = 0

class Object:
	pos = Point()
	ang = Point()


obj1 = Object()
obj2 = Object()

obj1.ang.x = 123

print obj2.ang.x    <<-- AND I FIND IT CONTAINS OBJ1's X VALUE
-------------------------------------------------------------------------



In C, this would be:

-------------------------------------------------------------------------
typedef struct Point {
	float  x;
	float  y;
	float  z;
} Point;

typedef struct Object {
	Point  pos;
	Point  ang;
} Object;

Object obj1;
Object obj2;

obj1.ang.x = 123;
printf("%f", obj2.ang.x);   <<-- AND NOW THE VALUES IN OBJ1 AND OBJ2 
				 WILL NOT BE RELATED (OBJ2 WILL JUST 
				 CONTAIN GARGAGE AT THIS POINT).
-------------------------------------------------------------------------


I've found that writing something like:

class Object(Point):
	...

works, but then I don't have a wrapper around the (x, y, z) values
anymore and I 
don't know how to bring in Point more than once. Is it possible to do
this in Python.  I've looked in the tutorial and Mark Lutz's book to no
avail.

Thanks,
Andy Beall



From bajandin at mail.ru  Wed Aug 25 02:51:19 1999
From: bajandin at mail.ru (Egor Bajandin)
Date: Wed, 25 Aug 1999 13:51:19 +0700
Subject: Tkinter documentation
References: <72805CBA8CD4D211A8090060945155F3063525D6@umkc-mail01>
Message-ID: <37C39267.1F1C325E@mail.ru>

"Cummings, Charles S. (UMC-Student)" wrote:

>
>
> Does anybody know where to get some end-all ultimate doc for Tkinter
> besides what's listed on python.org?
>
> Steven Cummings

Look at the www.pythonware.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990825/0bcd32ee/attachment.html>

From earlybird at mop.no  Sun Aug 29 17:07:24 1999
From: earlybird at mop.no (Alexander Staubo)
Date: Sun, 29 Aug 1999 23:07:24 +0200
Subject: PIL 1.0 binaries/makefiles on Win32 wanted
Message-ID: <MPG.1233bf735b72032698968b@news.mop.no>

Does anybody have compiled PIL 1.0 binaries for Win32 _or_ VC++ 
makefiles lying around?

-- 
Alexander Staubo             http://www.mop.no/~alex/
"`Ford, you're turning into a penguin. Stop it.'"
--Douglas Adams, _The Hitchhiker's Guide to the Galaxy_



From dtj at rincong.com  Sun Aug 29 00:02:14 1999
From: dtj at rincong.com (Derek Jones)
Date: Sat, 28 Aug 1999 21:02:14 -0700
Subject: Date Calculations
References: <37BF5D1A.9865CDF@fnmail.com>
Message-ID: <dtj-2808992102140001@ip-44-98.tus.primenet.com>

>I'm making some Date Calculaton class with things like
>deltaDays, addDays, daysInYear and so on, above the calendar 
>and time classes. After some 300 lines I thought, I better
>ask, whether that has already been done by someone else.
>How is this?

I am sure that the classes and libraries mentioned earlier in this thread
fit your bill more exactly.

However, for some simple, quick, classless date calculation, I have found
the standard 'time' module to be terrifically useful.  The big headache
(for me) with dates is handling things like a delta in days, or finding
out "what are all the dates of this week?", handling month rollover, leap
days, and other horrors.  The time.localtime and time.mktime functions are
really quite nice, and Python's tuple-unpacking makes the code quite
readable.  Attached as an example is a this_week() function that I needed
recently.  By always converting to and from the seconds-since-epoch that
the time module uses, date arithmetic becomes trivial.

--Derek

def this_week () :
    """
    Returns a list that includes every date this week, starting with
    Monday.
    """
    date_list = [ ]
    # We'll want all 5 weekdays for this week.
    for i in range (0, 5) :
        y, mo, d, h, m, s, wd, doy, dst = \
            time.localtime(time.time() - 86400*(weekday - i))
    date_str = "%2.2d/%2.2d/%4.4d" % (mo, d, y)
    date_list.append(date_str)
    return date_list

-- 
My real email address does not have a 'g' in it, for those who care.



From kistler at fnmail.com  Thu Aug 26 06:41:00 1999
From: kistler at fnmail.com (Per Kistler)
Date: Thu, 26 Aug 1999 12:41:00 +0200
Subject: Tkinter, intermediate loop?
References: <37C4EDF2.E7F2EB82@fnmail.com>
Message-ID: <37C519BC.5F64B303@fnmail.com>

Found it:-)
-Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From kranio at nospam.nospam.it  Tue Aug 31 11:09:05 1999
From: kranio at nospam.nospam.it (Kranio)
Date: Tue, 31 Aug 1999 15:09:05 GMT
Subject: calling python interpreter
References: <37cbccd8.5014784@news.tiscalinet.it>
Message-ID: <37ccefcf.1207310@news.tiscalinet.it>

I found a solution: chmod ugo+rx pippo.py. But ... should I do the
same for every script? Does exist a faster system?

Marco



From bkc at Murkworks.com  Wed Aug 11 12:52:39 1999
From: bkc at Murkworks.com (Brad Clements)
Date: Wed, 11 Aug 1999 12:52:39 -0400
Subject: How do I throw an exception class from C?
Message-ID: <7os9p5$q3o$1@news.clarkson.edu>

In my swig'd module  I can easily throw an exception string or tuple, but
how do I throw an exception class?

I want to create a new exception type that is a class, create a class object
with some init data, then set the exception for the current thread to that
object.

Easy in Python, hard in C. How do I do it?

Thanks,


--
Brad Clements,
bkc at murkworks.com





From pointal at lure.u-psud.fr  Fri Aug 20 06:49:14 1999
From: pointal at lure.u-psud.fr (Laurent POINTAL)
Date: Fri, 20 Aug 1999 10:49:14 GMT
Subject: Reading from serial port and dumping to console
References: <37BD20F8.9A984D7E@pc.jaring.my>
Message-ID: <37bd325c.257488729@news.u-psud.fr>

On Fri, 20 Aug 1999 09:33:44 +0000, Thomas Lim <englim at pc.jaring.my>
wrote:

>: t = open('/dev/tty0', 'w')

Try with t = sys.stdout

A+

Laurent.

---
Laurent POINTAL - CNRS/LURE - Service Informatique Experiences
Tel/fax: 01 64 46 82 80 / 01 64 46 41 48
email  : pointal at lure.u-psud.fr  ou  lpointal at planete.net 



From archer_sully at my-deja.com  Mon Aug 30 14:31:37 1999
From: archer_sully at my-deja.com (archer_sully at my-deja.com)
Date: Mon, 30 Aug 1999 18:31:37 GMT
Subject: Bit fields
Message-ID: <7qeim8$3su$1@nnrp1.deja.com>

Hi,

I'm writing some extensions to set up a general SCSI oriented
test platform in Python.  Obviously there's a certain amount
of C work that has to be done, but I would like to do one of the
following:

A. access bit fields from a C struct that is returned to Python
as a string

or

B. encapsulate the structs as classes with all of the methods
implemented in C (this would appear to be possible but the
doc is mute).

A is preferable, because I wouldn't have to do write a new
class every time the firmware guys come up with a new struct
that is made up entirely of bit fields.  B is ok, because at
least I have emacs to generate the code for me.

Of course, I may be all wet with this question, in which case
I humbly ask for enlightenment ;-)

TIA,

Archer Sully
Software Engineer
OnStream, Inc.
<http://www.onstream.com>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fredrik at pythonware.com  Mon Aug 30 03:56:59 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 30 Aug 1999 07:56:59 GMT
Subject: Upper limit?!
References: <37CA1E0F.FA2A05A3@megsinet.net>
Message-ID: <014f01bef2bd$3fd09e90$f29b12c2@secret.pythonware.com>

David Oppenheimer <davidopp at megsinet.net> wrote:
> I was playing around with the command line and discovered that when I
> enter 2 to the 30th power, I get a number back...when I enter 2 to the
> 31st power I get an overflow error.  Is 2 to the 30th power near the
> upper limit for computations?!

just above it, on a 32-bit platform.  try using
use long integers instead, and compare the
result with the sys.maxint value:

>>> 2L**31
2147483648L
>>> import sys
>>> sys.maxint
2147483647

if you upgrade to an alpha, you'll get:

>>> 2**31
2147483648
>>> import sys
>>> sys.maxint
9223372036854775807

</F>




From aahz at netcom.com  Tue Aug  3 13:23:35 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Aug 1999 17:23:35 GMT
Subject: Any clues to source of this delay?
References: <al8vhaxtg1y.fsf@sirppi.helsinki.fi>
Message-ID: <7o78in$jmm@dfw-ixnews16.ix.netcom.com>

In article <al8vhaxtg1y.fsf at sirppi.helsinki.fi>,
Markus Stenberg  <mstenber at cc.Helsinki.FI> wrote:
>
>On Linux, single connections have 'glass ceiling' of 50 roundtrips/second;
>no matter what I kludge, it seems to stay there. Only way to fix this was
>to send 'empty' packets after processed ones, thus increasing throughput by
>about 5X.
>
>On NT, the glass ceiling is same, just 1/10 of Linux's (5
>roundtrips/second).

I can probably explain part of the NT problem: I bet you're using
Workstation instead of Server.  Workstation has artificial limits.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From dl231 at eng.cam.ac.uk  Wed Aug  4 13:11:34 1999
From: dl231 at eng.cam.ac.uk (D. Lopez-De-Ipina)
Date: Wed, 04 Aug 1999 18:11:34 +0100
Subject: Doubt about Pickle and CORBA IDL types
Message-ID: <37A87446.3640084B@eng.cam.ac.uk>

Hi,

I want to serialize some data of CORBA IDL types in a shelve. I have
just made a very simple program that serialises a CORBA.string and a
CORBA.short, the first one works but the second one not. Could anybody
please tell me why this happens?:

#!/usr/bin/env python

import shelve

# Fnorb modules.
from Fnorb.orb import CORBA

if __name__ == '__main__':
  dictionary = shelve.open('dummy')
  dictionary['str'] = CORBA.Any(CORBA.TC_string, 'hello')
  print "dictionary['str']:", dictionary['str'].value()

  dictionary['short'] = CORBA.Any(CORBA.TC_short, 5)
  print "dictionary['short']:", dictionary['short'].value()

[dl231 at worthingtons DynAny]$ python prueba.py
dictionary['str']: hello
dictionary['short']:
Traceback (innermost last):
  File "prueba.py", line 14, in ?
    print "dictionary['short']:", dictionary['short'].value()
  File "/lce/lib/python1.5/shelve.py", line 65, in __getitem__
    return Unpickler(f).load()
TypeError: ('too many arguments; expected 1, got 2', <class
Fnorb.orb.TypeCode.ShortTypeCode at 81705e0>, (tk_short,))

Thanks,

Diego




From m.faassen at vet.uu.nl  Sat Aug 21 20:57:15 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 22 Aug 1999 00:57:15 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <7ple6r$kf0@ds2.acs.ucalgary.ca> <37bec222@cs.colorado.edu>
Message-ID: <7pnhtb$mlm$1@newshost.accu.uu.nl>

Tom Christiansen <tchrist at mox.perl.com> wrote:
>      [courtesy cc of this posting mailed to cited author]
[snips]
> It's because @a is a first-class array, and $a merely a reference
> thereto.  Copying things not producing a copy is well, not what
> non-serious programmers immediately think of.  Why can python copy
> sequences of chars (1-byte strings) trivially:

>     s = "this stuff"
>     t = s

> But as soon as you have a sequence of, oh, integers for example, you
> can't use the same set-up.  Very odd.

Perhaps you misunderstand this. Python does not copy strings. Python only
makes a new reference to that string. But since that string is
immutable, it feels exactly the same as making a copy. But no actual
copy is made. Making a new reference to a completely immutable object is
functionally identical to making a copy of it.

The gotcha is this one: a tuple is an immutable object, but it can *contain*
mutable objects. So this rule doesn't apply for all tuples. But it does
for all strings.

Regards,

Martijn




From guy_oliver at yahoo.com  Sun Aug 29 11:56:31 1999
From: guy_oliver at yahoo.com (guy_oliver at yahoo.com)
Date: Sun, 29 Aug 1999 15:56:31 GMT
Subject: Suitable for Large Applications?!
References: <37C7F91A.B37535@megsinet.net>
Message-ID: <37c955ec.4000208@news.netset.com>

On Sat, 28 Aug 1999 10:58:35 -0400, David Oppenheimer
<davidopp at megsinet.net> wrote:

>What are the real performance issues.  When I hear someone say that
>copiling the Python code took hours and compiling equivalent C code took
>seconds, tends to get a newbie like me less enthused.  Should I take
>back my O'Reilly Python book and get a book on learning C++?!

Its been the case in most of my applications that the performance was
fine in python. Ive written many very large apps in python, but I find
that in general, my python programs are shorter than their counteparts
in C++ land.  

 But as the other poster mentioned, performance depends on what you
are trying to do.  If your writing gui's that get data from the
database, you'll probably never notice the lack in performance. Most
of my apps fall into the 'you cant tell' category.  But not all

 If, on the other hand, you are trying to do very large simulations or
calculations, then it may be best to extract those bits from the
python code and implement them as an extension module.  you still get
all the benefits of using python at the high level, and you get the
benefits of C's performance at the low level.  Best of both worlds, so
to speak.  Same goes for tight integration with hardware, like
hardware drivers, etc.  Those are best written in C or someother
compiled language, but you can still wrap those drivers in python
modules so you can use the hardware from inside python.

I'd say learning them both is probably the best idea.  I'm much more
productive in Python than I am in C, yet there are times where I can
not get the behavior I want out of python, and am forced to extend it
in C.  Its not as hard as it sounds, and the positives outweigh the
negatives.  

-Guy



From rodzilla2000 at my-deja.com  Sat Aug  7 12:03:57 1999
From: rodzilla2000 at my-deja.com (rodzilla2000 at my-deja.com)
Date: Sat, 07 Aug 1999 16:03:57 GMT
Subject: Python GUI
References: <01bee0af$3cdbf7e0$LocalHost@para.cis.scu.edu.tw>
Message-ID: <7ohld9$5ht$1@nnrp1.deja.com>

In article <01bee0af$3cdbf7e0$LocalHost at para.cis.scu.edu.tw>,
  "Para Hsu" <para at o3.net> wrote:
>
> -----???l?l??-----
> ?H????: Kevin Dahlhausen <morse at harborcom.net>
> ?s?D?s??: comp.lang.python
> ??????: python-list at python.org <python-list at python.org>
> ????: 1999?~8??6?? PM 08:19
> ?D??: Re: Python GUI
>
> >There is a TK/TCL builder called 'SpecTCL.' that has a Python module
called
> >"SpecPython."  I don't have the URL though.   It works very well, if
you
> like

I found a reference to an ftp site:
ftp://ftp.python.org/pub/python/contrib/Graphics/Tkinter/SpecPython0.2.t
gz

havent' tried to see if it works yet...
>
>     Would anybody tell me where the SpecPython is ?
>     I just can't find it using search engine, but SpecTCL only.
>     I've noticed some discussion of it. But no URL there.
>     Thanks in advance.
>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mnot at mnot.net  Fri Aug 20 22:20:40 1999
From: mnot at mnot.net (Mark Nottingham)
Date: Sat, 21 Aug 1999 12:20:40 +1000
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu> <37BCA74C.D3DCF5DD@fnmail.com> <3bgv3.759$mV.257016@news.shore.net>
Message-ID: <093501beeb7b$c37ff400$0301a8c0@mnot.net>

Not as a perldoc-ish sort of thing, but IMHO it would be helpful to have a
standard of documentation and installation for extention modules (as well as
library add-ons).


> Per Kistler <kistler at fnmail.com> wrote:
> : Tom Christiansen wrote:
>
> :> The fact that they did icky windows help and not man pages says
> :> something significant.
>
> : Why not create something like perldoc for python?
> : There must be others missing it...
>
> This is a big "not me".
>
> Missing a manpage?  At times, but I've got the info files, the O'Reilly
> quick reference, local copy of the HTML docs (thru lynx or netscape),
> etc.
>
>   -Arcege
>
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From hinsen at cnrs-orleans.fr  Tue Aug 31 05:08:51 1999
From: hinsen at cnrs-orleans.fr (Konrad Hinsen)
Date: 31 Aug 1999 11:08:51 +0200
Subject: Suitable for Large Applications?!
References: <37C7F91A.B37535@megsinet.net>
Message-ID: <m33dx0iafw.fsf@chinon.cnrs-orleans.fr>

David Oppenheimer <davidopp at megsinet.net> writes:

> apllications.  On one hand I hear people discussing applications that
> have 10,000 plus lines and on the other I hear people say its really
> only good for small tasks and for modeling.  I'd hate to spend time

I doubt that the size of project, counted in lines or whatever else,
is the main criterion to decide whether Python is appropriate or not.
The complexity of data structures and algorithms, the size of data
objects, the number of people working on the project, etc., are
important as well.

But I can't think of a case in which Python would be of no benefit at
all. I'd expect every big project to end up with a mixture of Python
and C/C++ code, with the amount of C/C++ code strongly depending on
details of the project.

My own big project, the Molecular Modelling Toolkit
(http://starship.python.net/crew/hinsen/mmtk.html), is currently at
12000 lines of Python plus 7000 lines of C. And I don't even want to
imagine how big it would be in pure C; it's a purely academic question
since most certainly I couldn't have developed it in pure C.

So my suggestion is: go ahead with Python, but be prepared to add
C extension modules. The combination is *much* more pleasant than
pure C or C++.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------



From aaron_watters at my-deja.com  Thu Aug 19 14:24:34 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Thu, 19 Aug 1999 18:24:34 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <slrn7ro8n1.h6v.jblaine@shell2.shore.net> <37bc29d8@cs.colorado.edu>
Message-ID: <7phi3v$95d$1@nnrp1.deja.com>

Greetings tchrist.  Hope to shake hands at the ORA conference
soon.

In article <37bc29d8 at cs.colorado.edu>,
  tchrist at mox.perl.com (Tom Christiansen) wrote:
> I keep thinking that everything should work on
> everything.  For example, this surprises me:
>
>     x = {1:2, 3:4}
>     y = {10:20, 30:40}
>     z = x + y
>
> The latter isn't legal.  But in Perl, you can merge hashes easily.
>
>     %x = (1,2,3,4);
>     %y = (10,20,30,40);
>     %z = (%x, %y);

To get exactly what you want you need my kjbuckets extension
( http://www.chordate.com/kjbuckets/index.html ),
but Python has D.update, viz:

>>> x = {1:2, 3:4}
>>> y = {10:20, 30:40}
>>> sum = {}
>>> sum.update(x); sum.update(y)
>>> sum
{30: 40, 3: 4, 10: 20, 1: 2}

But for my purposes this is not too useful since I normally
need to know after a dict union whether the union was ambiguous
(that is, caused key collisions)
or not -- hence the notion of a "dirty" or "clean" kjbuckets
structure.  I suspect the Perl version doesn't flag
ambiguities either, no?

Also, you might find some of the fancier kjbuckets
features (intersect, diff, composition, transitive closure,
reachability, etc) of use as well.
   -- Aaron Watters http://www.chordate.com

===

All generalizations are wrong.



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jwbaxter at olympus.net  Fri Aug 20 14:28:56 1999
From: jwbaxter at olympus.net (John W. Baxter)
Date: Fri, 20 Aug 1999 11:28:56 -0700
Subject: slice indices (RE: More random python observations from a perl programmer)
References: <001201beeacd$bd088500$712d2399@tim>
Message-ID: <jwbaxter-2008991128560001@otter.olympus.net>

In article <001201beeacd$bd088500$712d2399 at tim>, "Tim Peters"
<tim_one at email.msn.com> wrote:

> Python's flavor of slice notation will look just as obvious and natural when
> it "clicks" for you.

My reaction to slicing being "starting from...not including" was
"OK...I'll do it that way."  It didn't (and doesn't) seem like a big deal.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwb at olympus.net



From philh at vision25.demon.co.uk  Sun Aug  8 19:17:32 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 08 Aug 99 23:17:32 GMT
Subject: Python parser generators
Message-ID: <934154252snz@vision25.demon.co.uk>

What parser generators are available for Python? I'm current using 
PyLR, are there any others that people recommend?


-- 
Phil Hunt....philh at vision25.demon.co.uk




From robin at jessikat.demon.co.uk  Sun Aug 15 05:56:25 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 15 Aug 1999 10:56:25 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
 <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk> <37b610c1@cs.colorado.edu>
Message-ID: <I1oT+LAJ7ot3Ewv0@jessikat.demon.co.uk>

In article <37b610c1 at cs.colorado.edu>, Tom Christiansen
<tchrist at mox.perl.com> writes
>     [courtesy cc of this posting mailed to cited author]
>
>In comp.lang.python, 
>    Robin Becker <robin at jessikat.demon.co.uk> writes:
>:Lets face it; the perlers have a fairly hopeless position when it comes
>:to defending the perl language 
>
>I'm going to pretend you didn't just now so egregiously underestimate
>both the acumen and the tenacity of those whom you don't know, for
>any other response would lead inexorably to ratholes so tortuous as to
>captivate Borges, so black as to plunge Pollyanna into deepest despair.
>For your own personal benefit, however, you might in the future take
>some care to avoid repeating that particular stultiloquy. :-)
>
>--tom
defense defense, did you get upset from my reasonably positive post.
shame. I guess all these years you've been casting perls before this
swine. Flame on if that's what you like.
-- 
Robin Becker



From olivier.deckmyn at mail.dotcom.fr  Fri Aug 20 07:45:53 1999
From: olivier.deckmyn at mail.dotcom.fr (Olivier Deckmyn)
Date: Fri, 20 Aug 1999 13:45:53 +0200
Subject: Class Methods (as opposed to Instance Methods)
Message-ID: <7pjfhm$48p$1@feed.teaser.fr>

Hi,

I play with Python while a few now (1 month).

My programer background is ObjectPascal, C++ and SmallTalk.

There is an "object-feature" I am not able to reproduce with Python : class
method.
As opposed to instance methods, theese should apply to classes.
ex:

class MyClass:
    def myInstanceMethod(self):
        print "spam"

    def myClassMethod(?, value):
        if value=='1' :
            return 'MyClass'
        elif value=='2' :
            return 'Something'
        else: return 'What else?'

Instance methods are easy to use:

instance=MyClass()
instance.myInstanceMethod()

But here is how I would like to use class Methods :

print MyClass.myClassMethod(2)

(Note that there is no MyClass().xxx but well MyClass.xxx)

I don't find any way to do that :(
I know it exists class attributes, but it does not fit my needs :(

Please HEEEELP !

---
Olivier Deckmyn, Paris - France
"Any sufficiently advanced technology is indistinguishable from
magic." -Arthur C. Clark





From tim_one at email.msn.com  Fri Aug 20 03:30:11 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 20 Aug 1999 03:30:11 -0400
Subject: Compound Assignment Operators ( +=, *=, etc...)
In-Reply-To: <37bcb30b@cs.colorado.edu>
Message-ID: <001501beeadd$d69ae520$712d2399@tim>

[posted & mailed]

[Timothy R Evans]
> :x = 5
> :def foo():
> :    print x # works, prints 5
> :def bar():
> :    x = 6 # works, but doesn't change the global
> :def wibble():
> :    print x
> :    x = 6 # this fails with NameError
> :
> :Seems like a good behaviour as it stops you doing dumb things.

[Tom Christiansen]
> Perhaps it could give a better error message, pretty please?

I whined about this for years myself, and a month or two ago finally wrote a
patch to make it raise UnboundLocalError instead -- which is what you'll get
if you use the current CVS version of Python.  It should be made clear that
Timothy's wibble example fails on the "print x" line, though, not the
assignment.

> Sometimes driving Python feels like driving Ken Thompson's car.
> It just keeps saying> "?" and you're expected to know what
> that means.  :-(

But also like a sleek modern car with idiot lights, Python's syntax is so
simple you soon figure out that "?" means what it says <wink>.

change-the-oil-or-fix-the-syntax-ly y'rs  - tim





From linyong2000 at 990.net  Wed Aug 25 07:30:30 1999
From: linyong2000 at 990.net (Daniel Y.L.)
Date: Wed, 25 Aug 1999 19:30:30 +0800
Subject: Good book for using Python as CGI?
Message-ID: <7q0k7m$fs5$1@news.ntu.edu.sg>

Hi,

I was used to using Perl to write CGI programs.  Now, I am trying to study
Python and want to use Python to write my CGI programs.  I have a book
"Programming Python" by Mark Lutz.  But, is there any good books which can
teach me how to write CGI using Python?  Any good suggestions or useful web
sites?  Thanks a lot.







From bwarsaw at python.org  Thu Aug 12 21:34:50 1999
From: bwarsaw at python.org (bwarsaw at python.org)
Date: Thu, 12 Aug 1999 21:34:50 -0400 (EDT)
Subject: Zope for python.org?
References: <7ouecd$9m$1@nnrp1.deja.com>
Message-ID: <14259.30266.418140.625710@anthem.cnri.reston.va.us>

>>>>> "DCG" == Dinu C Gherman <gherman at my-deja.com> writes:

    DCG> Would it be desirable to turn python.org into a Python
    DCG> "portal site" using the increasingly popular Zope product?

I'm very interested in looking at Zopifying python.org at some point,
probably after Zope 2.0 is released.  If it helps better serve the
Python community by allowing a more direct distributed management of
the site, and coincidentally makes my job as webmaster easier, I'm all 
for it.

-Barry



From greg.ewing at compaq.com  Thu Aug 19 00:35:21 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 19 Aug 1999 16:35:21 +1200
Subject: Deepcopying numeric array
References: <37BA93CB.5050@cs.bham.ac.uk>
Message-ID: <37BB8989.7BA08CA3@compaq.com>

Stuart Reynolds wrote:
> 
> Does anyone know why the Numeric arrays can't be deepcopied? Is it just
> an omission?

Not only that, but I just found out a few minutes
ago that they're un-comparable and un-hashable!

Is there any good reason for these restrictions?
Are they on hold awaiting Rich Comparisons or
something?

Greg



From vijaym at netscape.com  Mon Aug  9 19:43:11 1999
From: vijaym at netscape.com (Vijay Mallela)
Date: Mon, 09 Aug 1999 16:43:11 -0700
Subject: internet programming with python
Message-ID: <37AF678E.CC8E165B@netscape.com>

Hi,
I am looking for the book "Internet programming with python". I need it
at the earliest..
I checked out amazon and other stores and they dont have it.. does any
one know where I can get the book
Thanks
vijay




From david.stegbauer at cz.opel.com  Tue Aug  3 03:22:58 1999
From: david.stegbauer at cz.opel.com (david.stegbauer at cz.opel.com)
Date: Tue, 3 Aug 1999 07:22:58 +0000
Subject: class Pivot
Message-ID: <C12567C2.002DD03E.00@derumg01.cyberlink.eds.com>


Hello,

below is code of my Pivot class. It is inspired by Excel "Pivot Tables" or
database "cross tables". Pivot act as very simple matrix with generalized
indexes - index can be any type with defined ordering (i.e. __cmp__ function in
class) and hash value (i.e. __hash__ function in class).

Pivot only allow setting and getting items and incrementing values. Note, that
if Pivot doesn't contain data for given indexes, it automatically append
required row/column and initially set new values to 0 (zero).

I'm using this class daily about two months in conjunction with my PivotWriter
class (still need to be improved). So Pivot seems to be functional, but it isn't
perfect.

8<-----------------------------------------------------
import string

#rows are continuous blocks
#r0c0, r0c1, r0c2, r1c0, r1c1, r1c2, etc...
class Pivot:
    """Pivot: matrix with any type of index and value (numeric prefered)
       1999 David Stegbauer, david.stegbauer at cz.opel.com
       freeware"""
    def __init__(self):
        self.rows = {}  #map row labels to row numbers
        self.rowcnt = 0 #row count
        self.cols = {}  #map column labels to numbers
        self.colcnt = 0 #column count
        self.data = []  #continuous block of own data
    def __del__(self):
        pass
    #---
    def _indexof(self, col, row):
        """calculate index to self.data from row and colum labels
        if such row or column is not present, it is added"""
        if self.rowcnt == 0:     #colcnt has to be also zero...
            self.cols[col] = 0
            self.rows[row] = 0
            self.data.append(0)
            self.rowcnt = 1
            self.colcnt = 1
            return 0
        if not self.cols.has_key(col):  #then append new column
            self.cols[col] = self.colcnt
            for i in xrange(self.rowcnt*self.colcnt, self.colcnt-1,
-self.colcnt):
                self.data.insert(i, 0)
            self.colcnt = self.colcnt + 1
        if not self.rows.has_key(row):  #then append new row
            self.rows[row] = self.rowcnt
            self.data = self.data + (self.colcnt) * [0]
            self.rowcnt = self.rowcnt + 1
        return self.rows[row] * self.colcnt + self.cols[col]    #calc index

    def setitem(self, col, row, value):
        """directly set value in Pivot"""
        i = self._indexof(col, row)
        self.data[i] = value

    def getitem(self, col, row):
        """get value from Pivot"""
        i = self._indexof(col, row)
        return self.data[i]

    def increment(self, col, row, val=1):
        """increment value in Pivot, default by one"""
        i = self._indexof(col, row)
        self.data[i] = self.data[i] + val

    def rownames(self):
        """get sorted row labels"""
        rn = self.rows.keys()
        rn.sort()
        return rn

    def colnames(self):
        """get sorted colum labels"""
        cn = self.cols.keys()
        cn.sort()
        return cn

    def coltotals(self):
        """get column totals (sums), bad ordered"""
        tot = (self.colcnt) * [0]
        for j in xrange(len(self.data)):
            i = j % self.colcnt
            tot[i] = tot[i] + self.data[j]
        # TO DO:  sort same as colnames() is
        return tot

    def formatrow(self, row, fmt, sep):
        if not self.rows.has_key(row):
            return ""
        cn = self.cols.keys()
        cn.sort()                   #column indexes
        i = self.rows[row] * self.colcnt   #index base
        r = ""                      #return value
        if type(fmt) != type("") or fmt == "":
            do_fmt = 0
        else:
            do_fmt = 1          #should apply format
        for e in cn:
            if do_fmt:          #apply format
                r = r + fmt % self.data[i + self.cols[e]] + sep
            else:
                r = r + `self.data[i + self.cols[e]]` + sep  #backticks - repr
        return r

    def format(self, itemformat = '', rowsep = '\n', colsep = '\t'):
        cn = self.cols.keys(); cn.sort()
        r =  reduce(lambda x,y,s=colsep: x+str(y)+s, cn, colsep)  #column labels
        cn = map(lambda i, dic=self.cols: dic[i], cn)             #column
indexes
        rnames = self.rows.keys(); rnames.sort()                  #row labels
        rn = map(lambda i, dic=self.rows: dic[i], rnames)         #row indexes
        r = r + rowsep

        if type(itemformat) != type("") or itemformat == "":
            do_itemformat = 0
        else:
            do_itemformat = 1          #should apply format
        i = 0
        for row in rn:
            r = r + str(rnames[i]) + colsep; i = i + 1
            for col in cn:
                if do_itemformat:          #apply format
                    r = r + itemformat % self.data[row*self.colcnt + col] +
colsep
                else:
                    r = r + str(self.data[row*self.colcnt + col]) + colsep
            r = r + rowsep
        return r





From kuncej at mail.conservation.state.mo.us  Mon Aug 30 12:14:53 1999
From: kuncej at mail.conservation.state.mo.us (Jeffrey Kunce)
Date: Mon, 30 Aug 1999 11:14:53 -0500
Subject: [HOWTO] Launching python from Windows Explorer
Message-ID: <s7ca67f3.082@mail.conservation.state.mo.us>

I've had several requests to elaborate on an earlier post.
Perhaps someone could check this, and add to an appropriate FAQ.
These procedures are for NT, there may be minor differences for Win9x.

To easily run/open python scripts from the windows explorer,
I define several "actions" for .py files. Once these actions
are defined on a machine, you can right-click on a python script,
and choose the action you want to happen.

Action Name: test
Application: "C:\Program Files\Python\python.exe" -i "%1"
Description: Runs the python script in an MSDOS window. Leaves
the window open, and the interpreter running, when the script
finishes. Good for debugging, esp when using the pdb module.

Action Name: wordpad
Application: "C:\Program Files\Windows NT\Accessories\wordpad.exe" "%1"
Description: Opens the python script for editing in wordpad.

Action Name: idle
Application: "C:\Program Files\Python\pythonw.exe" "C:\Program Files\Python\Tools\idle\idle.pyw" "%1"
Description: Opens the python script for editing/running in Idle. Each time
this action is used, it will open a new instance of Idle. To open additional scripts, 
use Idle's File/Open command.

To define an action for a ".py" files:
  -open Windows [NT] Explorer
  -from the "View" menu, select "Options"
  -select the "File Types" tab
  -from "Registered File Types", select "Python File" 
  -click the "Edit" button
  -click the "New" button
  -for "Action", enter the action name to be 
displayed in the popup menu
  -for "Application used", enter the command 
line (as in the examples above)
  -click "OK", etc...

Variations of this process will edit existing
actions, or define actions for new file types.

  --Jeff




From aahz at netcom.com  Tue Aug 17 14:38:22 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Aug 1999 18:38:22 GMT
Subject: Where to post job openings?
References: <roogg6miin.fsf@gage.scripps.edu>
Message-ID: <7pca6u$8ao@dfw-ixnews13.ix.netcom.com>

In article <roogg6miin.fsf at gage.scripps.edu>,
Don Bashford  <bashford at gage.scripps.edu> wrote:
>
>Is there a place to post Python-related job openings?  I don't find
>anyplace in the Python web sites.  Is posting to this newsgroup okay?

I've posted job openings here in the past and received no complaints.
You should definitely make the Subject: line clear, though, including
the geographic location.  E.g., "Subject: Python job opening (Mtn. View,
CA)" -- which happens to still be true....

I'd recommend not posting the ad too often, though it's always good to
be alert to opportunities (like this one, for me ;-) to casually mention
openings.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From amaranda at spider.com  Fri Aug 20 11:48:50 1999
From: amaranda at spider.com (Alex Maranda)
Date: Fri, 20 Aug 1999 16:48:50 +0100
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <37bc313b@cs.colorado.edu> <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us> <019301beeaea$004c4d80$f29b12c2@secret.pythonware.com> <37bd4876@cs.colorado.edu>
Message-ID: <37BD78E1.71F0@spider.com>

Tom Christiansen wrote:
> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.python,
>     "Fredrik Lundh" <fredrik at pythonware.com> writes:
> :as an aside, it's a bit sad that Tom didn't read the book's
> :preface, where Mark says, among other things:
> 
> It's a shame that so many of you should be so unbelievably harsh on me
> when *I'm* the one who with malice toward none and full of revulsion
> at the flames causes by *YOUR* community's ignorant attacks on mine,
Tom, you got it all wrong. That John whatever his name was who started
and fueled that stupid flame on c.l.p.m is not a representative of the
Python community by any stretch of the imagination. He ended up in the
killfile of most c.l.py newsgroup regulars. I've been reading this
newsgroup for 2 years now and it's the first time I've seen such violent
advocacy.

> that *I* made the honest effort to learn *YOUR* language, and that I
> furthermore for your benefit not my own then took the additonal trouble
> to write-up my experiences and observations gleaned while five hundred
> miles away from the nearest tree or internet link.
You got it all wrong again. You are WELCOME. It's just the style of your
posts - we're not used to it. You're definitely coming from a different
newsgroup culture. If you would just stay in O_RDONLY mode on c.l.py for
a couple of weeks you'll understand what I'm talking about.

> 
> And for that, I am abused for my candor and effort in trying to convey
> to you what that process was like for me.
Your effort is most welcome. Perhaps in time you will remove the bias
from your notes.

> 
> Shame on you!
[rest of prose snipped]

Regards,
Alex



From kent at tiamat.goathill.org  Fri Aug 27 19:04:47 1999
From: kent at tiamat.goathill.org (Kent Polk)
Date: 27 Aug 1999 23:04:47 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net> <slrn7s2lb3.k6b.abigail@alexandra.delanet.com> <7putsq$3ql$1@lycaeus.calstatela.edu> <MPG.122fcf84a811c6099896e8@news.melbpc.org.au>
Message-ID: <935793922.672565@fezzik.endicor.com>

On Fri, 27 Aug 1999 15:33:04 +1000, Factory wrote:
>
>  Hmm that might be a bit off, the anchient Greeks had no concept of 
>variables, thus you would never see f(x)= 2y in any old greek 
>mathematics. IIRC variables were discovered(?) by Al-Jabar sometime 
>after.

I believe you are thinking of Al'Khwarizmi, who wrote the book
"Hisab al-jabr w'al-muqabala". 'Algebra' came from 'al-jabr' and
'Algorithm' came from the name Al'Khwarizmi. Later Greeks certainly
appear to have used the concept of variables, but did have not the
algebraic notation.




From c.evans at clear.net.nz  Mon Aug  9 08:23:26 1999
From: c.evans at clear.net.nz (Carey Evans)
Date: 10 Aug 1999 00:23:26 +1200
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com> <37AA804D.9A1317C0@houbits.com> <37AEA5A6.5C556643@houbits.com>
Message-ID: <87907l5eip.fsf@psyche.evansnet>

Nigel Head <nhead at houbits.com> writes:

> Actually, having thought and done some research I feel like going for
> the Artistic License for any future stuff I may distribute. It doesn't
> prescribe commercial use but I've decided that was a Wrong Thing to
> want to do anyway. It does disallow selling the software (as opposed
> to using it commercially) which is what I really meant to say; apart
> from that it's Open Source certified and I can actually understand it
> -- as opposed to the GPL which reads too legalese for my small brain!

Well really, anything that is covered by the OSD _does_ allow selling
in some circumstances, which is what the Artistic license does in
section five.

If I decide that $100 is a reasonable copying fee, and whoever I sell
it to agrees or is convinced by my marketing, I can "sell" it for
$100.  (See the definitions.  Note that I don't have to justify to you
what I sell it for.)  Or I can use it in a $1000 package and sell
that, under section 8, as long as it isn't obvious I'm using your
software!

The Debian Free Software Guidelines, which the OSD is mostly based on, 
assume that a collection of Free Software can be sold for as much as
anyone will pay.  If you try to stop anyone from selling your software 
at all, it *won't* be Open Source.

-- 
	 Carey Evans  http://home.clear.net.nz/pages/c.evans/

	       "This is where your sanity gives in..."



From Friedrich.Dominicus at inka.de  Fri Aug 13 02:16:47 1999
From: Friedrich.Dominicus at inka.de (Friedrich Dominicus)
Date: Fri, 13 Aug 1999 08:16:47 +0200
Subject: portability question
Message-ID: <37B3B84F.F818F396@inka.de>

About file and directory handling on different platforms. 

I'm aware that I can read the source to find out. But maybe someone can
give me some other hints about that.

How is it done in Python? Do you think the Python solution of doing it
is ok?

Hope someone can give me some hints
Friedrich



From guido at cnri.reston.va.us  Thu Aug 19 14:52:53 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 19 Aug 1999 14:52:53 -0400
Subject: More random python observations from a perl programmer
References: <199908191713.LAA02152@jhereg.perl.com>
Message-ID: <5l4shvtxfe.fsf@eric.cnri.reston.va.us>

Tom Christiansen <tchrist at jhereg.perl.com> writes:

> I only have the Lutz book.  I don't have manpages. :-(

Tom, I've skimmed this thread and I think you have expressed your
desire for Python documentation formatted as man pages a more than
sufficient number of times.

The argument that there is no online documentation has been amply
refuted -- your only remaining complaint must be that they aren't
there in your preferred format.  Tough.

Please let the issue go.  As we say, you are welcome to fix it
yourself.  We'd even accept a Perl script that turns our latex (soon
to be SGML or XML) into man page format.

Since you're an O'Reilly author, I'm surprised you didn't base your
(otherwise excellent) Perl/Python guide on the much more recent (April
99) Learning Python by Mark Lutz and David Ascher.  It would have
saved you a whole bunch of pitfalls.

--Guido van Rossum (home page: http://www.python.org/~guido/)



From fredrik at pythonware.com  Sun Aug 22 12:16:00 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 22 Aug 1999 16:16:00 GMT
Subject: Need help with Tkinter
References: <7pof0u$av$1@nnrp1.deja.com>
Message-ID: <00c501beecbc$0f33c390$f29b12c2@secret.pythonware.com>

chinaski_py at my-deja.com wrote:
> When I try to import Tkinter from PythonWin this is the message I get:
> TclError: Can't find a
> usable init.tcl in the following directories:
>     {} ./lib/tcl8.0
> C:/tcl8.0/library {C:/Program Files/library}

http://www.pythonware.com/people/fredrik/fyi/fyi02.htm
might help.

</F>




From spamfranke at bigfoot.de  Mon Aug 16 09:02:07 1999
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Mon, 16 Aug 1999 13:02:07 GMT
Subject: L postfix char on longs - what is the best way to discard it.?
References: <934546409.21489.0.nnrp-14.d4e5b0e4@news.demon.co.uk>
Message-ID: <37b80b09.4885895@news.omnilink.de>

On Fri, 13 Aug 1999 13:22:22 +0100, "Jim Crumpler" <Jim.Crumpler at unicity.com.au> wrote:

>Hi folks - I'm batting to find a nice clean way to remove the 'L' from long
>repr strings .. (eg, my SQL queries end up like "SELECT * FROM blah WHERE
>guff = 23423L"..)
>
>I can fiddle with every bit of my code that attempts to display a long, such
>as
>
>query = "... WHERE guff = %s" % repr(long(x))[0:-1]
>
> -- its ugly when you use alot of longs, but does the job..
>
[snip]
>
>Does anyone have some suggestions on better methods?

Printing a long without the L-suffix would be a good candidate for an 
additional format specifier of the %-Operator IHMO.

Stefan




From tre17 at pc142.cosc.canterbury.ac.nz  Sun Aug 15 21:20:50 1999
From: tre17 at pc142.cosc.canterbury.ac.nz (Timothy R Evans)
Date: 16 Aug 1999 13:20:50 +1200
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov>
Message-ID: <c75yafced1p.fsf@pc142.cosc.canterbury.ac.nz>

Drew McDowell <drew.mcdowell at msfc.nasa.gov> writes:

> I looked in the FAQ and couldn't find anything on the Compound
> Assignment Operators.
> (+=, *=, -=, etc..)
> Why dosen't Python support them?  Is there a easy way to add them?
> 
> -Drew McDowell

One reason that these constructs could cause problems is due to
namespaces.  Assuming that x += y is expanded to x = x + y, consider
the following code while remembering how python treats namespaces and
global variables.

>>> def foo():
...    x += 2
...    return x
>>> x = 42
>>> foo()
44
>>> x
42

I don't think this is what you want to happen, but it is what python
would do.  Inside foo(), the global variable x is read, then a local
variable x is created (there is no `global x' in the function).  So in 
this case x += 2 does not add 2 to x, it creates a new local variable
that is 2 greater than the global variable of the same name.

Confused yet, well here's a second reason.

>>> x = [1,2,3,4]
>>> y = x
>>> x += [5,6]
>>> x
[1,2,3,4,5,6]
>>> y
what should this be?

you would expect x += y on lists to actually extend the list, or do
you want it to be x = x + y, which could be really slow on big lists,
while hiding the fact that it is copying the list.

--
Tim Evans





From dwelton at linuxcare.com  Thu Aug 12 15:44:59 1999
From: dwelton at linuxcare.com (David N. Welton)
Date: 12 Aug 1999 12:44:59 -0700
Subject: ifconfig C module
Message-ID: <87g11o941w.fsf@linuxcare.com>

I got permission to publish the results of my 'ifconfig' module.

Since it's rather small, I thought I would post the whole thing here.

I realize it's not perfect, but it does what I need.  I'll accept
patches to make it better.  I'm sure the license could be changed to
the Python one if there is interest in including it in the main
distribution.

-------------------------------

/* $Id: ifmodule.c,v 1.2 1999/08/12 19:43:42 dwelton Exp $ */

/* 
   Copyright (C) 1999 Linuxcare Inc.

   By David N. Welton <davidw at linuxcare.com>

   This program is free software; you can redistribute it and/or
   modify it under the terms of the GNU General Public License as
   published by the Free Software Foundation; either version 2 of the
   License, or (at your option) any later version.

   This program is distributed in the hope that it will be useful, but
   WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
   General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this program; if not, write to the Free Software
   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
   USA
 */

/* 
 To build under Debian GNU/Linux:

 gcc -shared -Wl,-soname,pyifconfig.so -o pyifconfig.so ifmodule.c
 */


#include <python1.5/Python.h>
#include <unistd.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <sys/ioctl.h>
#include <net/if.h>
#include <netinet/in.h>
#include <netinet/in.h>
#include <arpa/inet.h>

static PyObject *pyifconfig(PyObject *self, PyObject *args)
{
    int fd;

    struct ifreq ifreq;
    unsigned char *hw;
    struct sockaddr_in *sin;
    char *itf;

    char pyhwaddr[40];
    char pyaddr[20];
    char pybrdaddr[20];
    char pynetmask[20];

    if (!PyArg_ParseTuple(args, "s", &itf))
	return NULL;

    fd = socket(AF_INET, SOCK_DGRAM, IPPROTO_IP);

    strcpy(ifreq.ifr_name, itf);

    /* hardware address */
    ioctl(fd, SIOCGIFHWADDR, &ifreq);
    hw = ifreq.ifr_hwaddr.sa_data;
    sprintf(pyhwaddr, "%02x:%02x:%02x:%02x:%02x:%02x", 
	    *hw, *(hw + 1), *(hw + 2), *(hw + 3), *(hw + 4), *(hw + 5));
    
    /* address */
    ioctl(fd, SIOCGIFADDR, &ifreq);
    sin = (struct sockaddr_in *)&ifreq.ifr_broadaddr; 
    sprintf(pyaddr, "%s", inet_ntoa(sin->sin_addr));

    /* broadcast */
    ioctl(fd, SIOCGIFBRDADDR, &ifreq);
    sin = (struct sockaddr_in *)&ifreq.ifr_broadaddr; 
    sprintf(pybrdaddr, "%s", inet_ntoa(sin->sin_addr));

    /* netmask */
    ioctl(fd, SIOCGIFNETMASK, &ifreq);
    sin = (struct sockaddr_in *)&ifreq.ifr_broadaddr; 
    sprintf(pynetmask, "%s", inet_ntoa(sin->sin_addr));

    close(fd);
    return Py_BuildValue("{s:s,s:s,s:s,s:s}", 
			 "hwaddr", pyhwaddr, 
			 "addr", pyaddr, 
			 "brdaddr", pybrdaddr, 
			 "netmask", pynetmask);
}

/* The command pyifconfig.pyifconfig(interface_name), where
   interface_name is the name of the interface, such as 'eth0' or 'lo'
   returns a dictionary with the following keys:

   hwaddr: hardware address
   addr: ip address
   brdaddr: broadcast address
   netmask: netmask
 */

static struct PyMethodDef pyifmethods[] = {
    {"pyifconfig", pyifconfig, METH_VARARGS},
    {NULL, NULL}
};

void initpyifconfig()
{
    Py_InitModule("pyifconfig", pyifmethods);	
}


------------------------------

Have fun,
-- 
David N. Welton, Web Engineer, Linuxcare, Inc.
415.354.4878 x241 tel, 415.701.7457 fax
dwelton at linuxcare.com, http://www.linuxcare.com/
Linuxcare. At the center of Linux.



From fdrake at acm.org  Fri Aug  6 10:17:38 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 6 Aug 1999 10:17:38 -0400 (EDT)
Subject: [Image-SIG] Proper application of the buffer interface
In-Reply-To: <37AAA7FE.FFE0FD2@lemburg.com>
References: <Pine.WNT.4.05.9908042019270.40-100000@david.ski.org>
	<14249.43677.175790.689266@weyr.cnri.reston.va.us>
	<37AAA7FE.FFE0FD2@lemburg.com>
Message-ID: <14250.61058.35328.260812@weyr.cnri.reston.va.us>

M.-A. Lemburg writes:
 > Wait... what's so bad about buffer(struct.pack()) ? Strings already
 > know the buffer interface, so this works just fine already.

  I think the issue is that if strings ever change to Unicode, we
don't want non-character data to ever be exposed as a string.  Using a 
buffer takes care of this and works now, allowing preparations for
further integration of Unicode to be tested gradually, rather than
having a complete meltdown from a large cutover.
  Functions which need to receive binary data will always use s#,
because s will raise TypeError if '\0' is in the string.
  The other way around is done less consistently now. The S format
character might not raise an exception if '\0' is in the data, and may 
standard functions and methods probably need to be adjusted to finish
up a Unicode integration, but I think using a buffer object for binary 
data will only avoid problems, not create them.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From krodgers at tdyryan.com  Mon Aug 23 12:28:21 1999
From: krodgers at tdyryan.com (krodgers at tdyryan.com)
Date: Mon, 23 Aug 1999 16:28:21 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com>
Message-ID: <7prsqk$9gr$1@nnrp1.deja.com>

In article <014101beecc0$e810e590$f29b12c2 at secret.pythonware.com>,
  "Fredrik Lundh" <fredrik at pythonware.com> wrote:
> on the other hand, I'm not sure that would have helped.
> judging from most threads that tchrist appears in (be it
> on comp.lang.perl.misc, gnu.misc.discuss, or this group),
> things often turn into a great flame fest when he's in-
> volved, by some mysterious reason.
>
> maybe we should add a note to the newsgroup FAQ?
>
> </F>

The comp.lang.fortran FAQ once had (maybe still has; I haven't looked at
it in a while) an entry titled something like "Why are there useless
debates?" that centered on a certain person at Purdue . . .
--
Kevin Rodgers  Northrop Grumman
"Snakes.  Why did it have to be snakes???" -- Indiana Jones


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From skip at mojam.com  Thu Aug 26 12:46:42 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 11:46:42 -0500 (CDT)
Subject: Returning none
In-Reply-To: <37C5683D.91D61FD0@rubic.com>
References: <37C55044.8CA02530@prescod.net>
	<37C5683D.91D61FD0@rubic.com>
Message-ID: <14277.28442.44107.591412@dolphin.mojam.com>

    Jeff> I think this proposal would break too much existing code, even for
    Jeff> a Python 2 change.

	...

    Jeff> Paul Prescod wrote:

    >> I think it would be more pythonish if this would return a runtime 
    >> (or even compiletime) error:
    >> 
    >> def foo():
    >> a=5
    >> 
    >> b=foo()
    >> 
    >> It shouldn't just return None. That's a source of errors.  This is
    >> obviously an incompatible change so I propose it for Python 2.

How about a compromise?  Perhaps Aaron Watters' PyLint can be modified to
bitch about functions without a return as the last statement?

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From vaboy65 at aol.com  Tue Aug 17 02:38:56 1999
From: vaboy65 at aol.com (vaboy65 at aol.com)
Date: Tuesday, 17 Aug 1999 00:38:56 -0600
Subject: Hot Talk 92621
Message-ID: <17089900.3856@aol.com>

Want to talk with REAL horny girls?

This is something VERY new in phone sex. The women on the line are NOT paid. They call for free phone sex with guys like YOU!

Guys: 50 cents per minute

Girls: TOTALLY FREE

=====================>1-888-335-HOTT (toll-free)

=====================>1-888-302-ORGY (toll-free) You can be billed discreetly directly to your phone bill.
-----------------------------------------------------------------

For gay talk call:

1-888-800-GUYS (toll-free)

-------------------------------------------------------------------

For Dominance & Submission Call:

888-700-WHIP

---------------------------------------------------------------------


4u#P;7,L



From jstevens at basho.fc.hp.com  Fri Aug 13 23:07:09 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:07:09 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132014.OAA03479@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:14 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cf4d@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :I had thought the arguments re: indentation and use of white space had
> :been settled years
> :ago. . .
> 
> Don't be silly.  We merely contained our laugher for the nonce.

Perhaps you should read the relevant papers from the DOD studies, the
human perception types, and the language theory types. . .

Seems pretty proven, to me.

John S.




From skip at mojam.com  Tue Aug 17 11:18:04 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 17 Aug 1999 10:18:04 -0500 (CDT)
Subject: if you want POSIX, import posix, not os
In-Reply-To: <1Feu3.381$mV.141087@news.shore.net>
References: <37B8EC03.835F8A51@w3.org>
	<14265.5096.537800.149227@buffalo.fnal.gov>
	<37B91E9A.2451936C@w3.org>
	<001301bee891$82fccc60$f29b12c2@secret.pythonware.com>
	<1Feu3.381$mV.141087@news.shore.net>
Message-ID: <14265.31990.925778.245319@dolphin.mojam.com>

    Michael> The better question is:
    Michael>    Why did it change from documenting these components in the OS-
    Michael>    dependant sections (specifically only posix) to documenting them
    Michael>    mostly in the module-os section in the current release?

I think it was because they were trying to steer people toward

    import os

so new users didn't know to look in the posix chapter for the docs.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From fleck at informatik.uni-bonn.de  Sun Aug 22 03:01:24 1999
From: fleck at informatik.uni-bonn.de (Markus Fleck)
Date: 22 Aug 1999 07:01:24 GMT
Subject: [ANNOUNCE] M2Crypto 0.01
References: <37bd6fa4.0@news.cyberway.com.sg> <37BED351.55E7DA98@lemburg.com> <37bee546.0@news.cyberway.com.sg>
Message-ID: <7po784$lpu@news.rhrz.uni-bonn.de>

Ng Pheng Siong <ngps at madcap.dyn.ml.org> wrote:
>> Could you please tell me where that server is located and whether
>> it is legal to download the software from Germany ? 
> The server is located in Singapore. AFAIK, there is no law against
> export of cryptography from Singapore.

That seems to be correct. But (oddly enough),
Singapore has *import* restrictions on cryptography:

  http://cwis.kub.nl/~frw/people/koops/cls2.htm#si ("Crypto Law Survey")

Yours,
Markus.

-- 
////////////////////////////////////////////////////////////////////////////
   Markus B Fleck - University of Bonn - CS Department IV - WHOIS MF5079
          UNIX Administrator - comp.lang.python.announce Moderator
     "GNU Gather" Free Internet Groupware Project - http://gather.net/
\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\



From greg.ewing at compaq.com  Thu Aug 19 00:34:06 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 19 Aug 1999 16:34:06 +1200
Subject: Deepcopying numeric array
References: <37BA93CB.5050@cs.bham.ac.uk>
Message-ID: <37BB893E.FD42CA0B@compaq.com>

Stuart Reynolds wrote:
> 
> Does anyone know why the Numeric arrays can't be deepcopied? Is it just
> an omission?

Not only that, but I just found out a few minutes
ago that they're un-comparable and un-hashable!

Is there any good reason for these restrictions?
I want to use one as a dictionary key...

Greg



From cgw at fnal.gov  Tue Aug 17 15:39:48 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 17 Aug 1999 19:39:48 GMT
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <14265.29899.470021.309355@dolphin.mojam.com>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
Message-ID: <14265.47748.241386.291144@buffalo.fnal.gov>

Ole-Kristian Villaboe <olevi at my-deja.com> wrote:
 > 
 >     Ole> I've been writing an algorithm myself, but it's incredibly slow and
 >     Ole> I would like something quicker (of course :) )
 > 

To which Stephan Houben <stephan at pcrm.win.tue.nl> replied:

> 
> def randomize(l):
>     length = len(l)
>     for i in range(length):
>         j = rand.randrange(i, length)
>         l[i], l[j] = l[j], l[i]
> 
> Note that this randomizes the list in-place.

and Skip Montanaro shared:
 > Got this off the list a couple years ago.  Does the trick for me:
 > 
 >     def shuffle(lst):
 > 	"randomize a list"
 > 	for i in range(len(lst)-1,0,-1):
 > 	    choice = int(whrandom.random() * i)
 > 	    lst[choice], lst[i] = lst[i], lst[choice]
 > 
 >     >>> l = range(20)
 >     >>> shuffle(l)
 >     >>> l
 >     [4, 17, 12, 6, 7, 15, 1, 11, 16, 10, 5, 3, 9, 19, 0, 8, 18, 13, 14, 2]
 > 

and Jeremy Hylton chimed in with:

> I've rather fond of the following approach:
>
> import random
> 
> def randomize(a, b):
>    return random.choice([-1, 0, 1])
>
> l = range(2, 20)
> l.sort(randomize)


If you're playing poker with Skip, Jeremy, Stefan, you better let
Stefan shuffle the cards!!  The other two methods posted above are
seriously flawed.

By the way, Skip, I know you didn't write the "shuffle" method, just
passed it along, but in the following I'll call it "Skip's Method"
since I don't know who the original author is.  No criticism implied.


If I may be excused an excursion into group theory, the set of
permutations of a set of N elements forms a group, called the
symmetric group, Sym(N).  The symmetric group has N! elements, and
Ole's question can be interpreted as "How do I pick randomly
distributed elements of Sym(N)".

There is a function defined on Sym(N) called "sign" which is the sign of 
a permutation - any permutation can be undone by a series of "flips"
(e.g. l[i],l[j]=l[j],l[i] where i!=j) and for any permutation, the
number of "flips" required is either odd or even (the number itself is 
not determined by the permutation, because you could always add 2 more 
flips that just undo each other, but the *sign* of the number is).

For a permutation p, define sign(p) as (-1)**number of flips to undo
the permutation.  (This is equal to the determinant of the matrix
representing p, a matrix which only contains 1's and 0's).

Now, the group Sym(N) has an even number of elements and exactly half
have sign 1 (even permutations) and half have sign -1 (odd permutations).

Thus, 

Integral over (p in Sym(N)) of sign(p) == 0

Furthermore, if P acts on the vector [0,1,...,N-1] then we have

Integral over (p in Sym(N)) of p[0] = (N-1)/2 
	(the mean of 0,1,...N-1, since we have taken all permutations)

If we have a truly random way of choosing elements of Sym(N) we can do a
Monte Carlo integral, which should approximate the acutal integral,
which is just a finite sum in this case.

We can use this to test the posted solutions:

#!/usr/bin/env python

import math

#Stefan's method
import random; rand=random
def randomize(l):
    length = len(l)
    for i in range(length):
        j = rand.randrange(i, length)
        l[i], l[j] = l[j], l[i]


#Skip's method
import whrandom
def shuffle(lst):
    for i in range(len(lst)-1,0,-1):
        choice = int(whrandom.random() * i)
        lst[choice], lst[i] = lst[i], lst[choice]

#Jeremy's method        
def jeremy(a, b):
    return random.choice([-1, 0, 1])

def hylton(seq):
    seq.sort(jeremy)
    

#calculate the sign of the permuted sequence seq
# seq is assumed to be a permutation of the list 0..N-1
def sign(seq):
    sgn=1
    seq=seq[:] #copy the sequence
    for i in range(len(seq)):
        if seq[i]!=i:
            j=seq.index(i)
            sgn = -1*sgn
            seq[i],seq[j]=seq[j],seq[i]
    return sgn



# Take the sequence 0..seqlen-1 and take
# ntries permutations chosen according to method,
# keep track of the means and standard deviations
# of the permuted elements  and also the mean 
# of the sign of the permutation
def means(seqlen, ntries, method):
    mean_sign = 0
    means = [0]*seqlen
    mean_square = [0]*seqlen
    for x in range(ntries):
        seq=range(seqlen) #sorted sequence
        method(seq)
        mean_sign = mean_sign + sign(seq)
        for i in range(seqlen):
            j=seq[i]
            means[i]=means[i]+j
            mean_square[i]=mean_square[i]+j*j
    ntries=float(ntries)
    for i in range(seqlen):
        means[i]=means[i]/ntries
        mean_square[i]=math.sqrt(mean_square[i]/ntries - means[i]*means[i])
    mean_sign=mean_sign/ntries
    return means, mean_square, mean_sign
   
for method in [randomize, shuffle, hylton]:
    print "Testing", method
    for seqlen in (2,3,4,5,6,7,8):
        mean, rms, mean_sign = means(seqlen, 50000, method)
	print "len=",seqlen
        print " mean values=", mean
        print " rms=", rms
        print " mean sign=",mean_sign


And here's the output:

Testing <function randomize at 811c208>
len= 2
 mean values= [0.50238, 0.49762]
 rms= [0.499994335568, 0.499994335568]
 mean sign= -0.00476
len= 3
 mean values= [1.00074, 0.99718, 1.00208]
 rms= [0.816908472474, 0.81641413976, 0.816159098216]
 mean sign= 0.0064
len= 4
 mean values= [1.4945, 1.49814, 1.49742, 1.50994]
 rms= [1.11800257155, 1.11937328019, 1.11683183318, 1.11786456979]
 mean sign= 8e-05
len= 5
 mean values= [1.98744, 2.00134, 2.0009, 2.00898, 2.00134]
 rms= [1.41024900156, 1.41502586704, 1.4200208414, 1.40725952106, 1.41838577418]
 mean sign= -0.00088
len= 6
 mean values= [2.49602, 2.51312, 2.49238, 2.50368, 2.48962, 2.50518]
 rms= [1.71489479549, 1.70544652968, 1.7037787226, 1.7034043729, 1.7096819165, 1.70960029469]
 mean sign= 0.0006
len= 7
 mean values= [2.9922, 2.99634, 2.99654, 3.00452, 3.01642, 2.99032, 3.00366]
 rms= [1.99693243752, 1.9949753393, 1.99988200362, 2.00073475743, 2.00253598809, 2.00430194771, 2.0005015882]
 mean sign= 0.00636
len= 8
 mean values= [3.4883, 3.50384, 3.5003, 3.50828, 3.49524, 3.50056, 3.49318, 3.5103]
 rms= [2.29326472741, 2.29340472974, 2.28332212138, 2.29005053254, 2.29190255953, 2.2880121692, 2.29673539782, 2.29349818182]
 mean sign= -0.00124

Testing <function shuffle at 811c228>
len= 2
 mean values= [1.0, 0.0]
 rms= [0.0, 0.0]
 mean sign= -1.0
len= 3
 mean values= [1.49934, 1.00132, 0.49934]
 rms= [0.4999995644, 0.9999991288, 0.4999995644]
 mean sign= 1.0
len= 4
 mean values= [1.99896, 1.65842, 1.34248, 1.00014]
 rms= [0.817825726179, 1.24588245979, 1.2511384614, 0.81538946547]
 mean sign= -1.0
len= 5
 mean values= [2.50058, 2.24862, 1.99542, 1.75502, 1.50036]
 rms= [1.11721066214, 1.48050264964, 1.58308528627, 1.47751304549, 1.11744345289]
 mean sign= 1.0
len= 6
 mean values= [3.00468, 2.80428, 2.61678, 2.37518, 2.19412, 2.00496]
 rms= [1.41707377987, 1.71985280812, 1.85314932793, 1.8534562222, 1.71629759238, 1.41553360907]
 mean sign= -1.0
len= 7
 mean values= [3.50838, 3.3154, 3.17066, 3.01972, 2.81974, 2.6719, 2.4942]
 rms= [1.70175491056, 1.97085840181, 2.11436400944, 2.15678722214, 2.11953917926, 1.97356793397, 1.71074438769]
 mean sign= 1.0
len= 8
 mean values= [4.00358, 3.85536, 3.711, 3.55926, 3.42784, 3.29538, 3.13716, 3.01042]
 rms= [1.99722987751, 2.23899961376, 2.36665988262, 2.43745117949, 2.44034279035, 2.37833779258, 2.23953279378, 1.99552284467]
 mean sign= -1.0

Testing <function hylton at 811c268>
len= 2
 mean values= [0.1123, 0.8877]
 rms= [0.315735189676, 0.315735189676]
 mean sign= 0.7754
len= 3
 mean values= [0.22454, 0.99842, 1.77704]
 rms= [0.546773982922, 0.498013557647, 0.497884362478]
 mean sign= 0.62736
len= 4
 mean values= [0.42664, 1.2663, 1.8238, 2.48326]
 rms= [0.901874886223, 0.890788588836, 0.753467690084, 0.739621370973]
 mean sign= 0.40748
len= 5
 mean values= [0.53772, 1.39702, 2.1733, 2.62146, 3.2705]
 rms= [1.08333614432, 1.08555751557, 1.0958955744, 0.991931181282, 0.964764090335]
 mean sign= 0.28956
len= 6
 mean values= [0.66926, 1.58918, 2.65972, 2.66882, 3.29658, 4.11644]
 rms= [1.31366321879, 1.37545880622, 1.51967414981, 1.18339334441, 1.18078800112, 1.13828016165]
 mean sign= 0.15808
len= 7
 mean values= [0.85156, 1.8575, 2.74986, 3.1399, 3.303, 4.0936, 5.00458]
 rms= [1.61084001887, 1.74200853901, 1.70962275968, 1.61907627677, 1.38761341879, 1.38874009087, 1.28955768525]
 mean sign= 0.1106
len= 8
 mean values= [1.064, 2.17124, 2.95968, 3.61052, 3.5423, 4.21624, 4.80906, 5.62696]
 rms= [1.93729295668, 2.13704395425, 2.00799758406, 2.03430217264, 1.72304692623, 1.7540240199, 1.61719569515, 1.44652727537]
 mean sign= 0.07044


The "randomize" function has good behavior - the sign of the
permutations generated averages out to 0, and the means and standard
deviations are close to the values we'd expect if the permutations
mixed up the indexes thoroughly and without bias.  I think this code
is good.

The "shuffle" function generates only odd or even permutations,
depending on the sign of n.  This is because at each step of the
iteration, two elements are *always* exchanged.  The means are
monotonically increasing (the permutation tends to produce increasing
sequences) oand the RMS variation is not equidistributed.  It peaks
(apparently) near the middle.

The "hylton" method favors even permutations over odd ones, tends to
produce increasing sequences, and distributes the variance unequally,
apparently peaking near the beginning (1/e of the way through?).  I
haven't quite fully analyzed the reasons for this yet...


Use-Stefan's-method-ly yr's,

		       Charles



From bowman at montana.com  Sat Aug 21 00:02:40 1999
From: bowman at montana.com (bowman)
Date: Fri, 20 Aug 1999 22:02:40 -0600
Subject: Man pages would be cool.
References: <37BC6986.A66B044E@starmedia.net>
			<199908191634.KAA01333@jhereg.perl.com>
			<14268.21198.934117.241668@anthem.cnri.reston.va.us>
			<14268.23007.239507.166209@weyr.cnri.reston.va.us> <y0j7lmq44wd.fsf@vier.idi.ntnu.no> <37BDD1B9.99588CB@prescod.net>
Message-ID: <37BE24E0.79EA064E@montana.com>

Paul Prescod wrote:
> 
> I was very distressed to load up my Red Had system and find Python
> effectively invisible. It should be available in man pages and IDLE
> should be in the GUI menus. (obviously two very different problems)

I'll pass on the man pages noise, but it is easy enough to add IDLE to
the menus. In fact, it fits very nicely in the slot where xeyes used to
live :)

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From carlos.ors at usa.net  Thu Aug 12 12:19:34 1999
From: carlos.ors at usa.net (Carlos Kyoku)
Date: Thu, 12 Aug 1999 16:19:34 GMT
Subject: A doubt of jpython
Message-ID: <19990812161934.8243.qmail@nwcst091.netaddress.usa.net>

Hi!

I have a problem with jpython. I do a class in java that has native methods. I
load jpython and import this class and, when I call the native
method give me an error. 

The general question is: can I use a java class that has natives methods
inside jpython?

                              A lot of thanks.



From tim_one at email.msn.com  Sun Aug  8 13:19:17 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 8 Aug 1999 13:19:17 -0400
Subject: Problem with modules refering to each other
In-Reply-To: <934111569snz@vision25.demon.co.uk>
Message-ID: <000701bee1c2$25b868a0$722d2399@tim>

[Phil Hunt, mutual imports]

Have you looked at the FAQ?  4.37, "How can I have modules that mutually
import each other?".

the-answer-is-"carefully"-ly y'rs  - tim





From fredp at multimania.com.nospam  Tue Aug 31 10:57:21 1999
From: fredp at multimania.com.nospam (Fred Pacquier)
Date: 31 Aug 1999 14:57:21 GMT
Subject: Newbie (read: pot. stupid) question on objects/namespaces
References: <8E339DC2FPaCmAnRDLM@news.dial.oleane.com> <m3iu5w5an5.fsf@pcrm.win.tue.nl>
Message-ID: <8E33ADDF0PaCmAnRDLM@news.dial.oleane.com>

stephan at pcrm.win.tue.nl (Stephan Houben) said :


>Let me try to rephrase the question first:
>You want to know if it is possible to find out, given a class or
>instance variable, to get at the class or instance.

Yes, that is a much shorter and more correct formulation !

>Short answer: "no".

Hmm. Somehow I felt that one coming :-)

>Longer answer: 
>This is really a tricky problem what "class or instance variable"
>actually means in Python. If you have a class like this:
>class Foo:
>    x = 1
>, then Foo.x refers to the number 1, which is an object in Python.
>Many other classes Bar, Baz, and instances of those, might have
>references to the same object 1. So 1 doesn't really have a single
>"owner". 

Understood.

>Now, you could have a dedicated object which registers its official
>"owner". Say you do that in this way:
>class MyClass:
>    def __init__(self, owner):
>        self.owner = owner
>
>You create instances of this object like this:
>class Bar:
>    x = MyClass(Bar)
>
>, or, if you want to have the instance, rather than the class, as owner:
>
>class Baz:
>    def __init__(self):
>        self.x = MyClass(self)

Yes, that is indeed very close to the result I was looking for. I might 
even have stumbled upon something similar eventually, without ever 
realizing that:

>Note that in *both* cases you create a circular reference which means
>that CPython cannot reclaim the class Bar or the instance of Baz.

Duh. Is this a total "no-no", or what ? If this is only for a handful of 
instances that are supposed to persist through the entire program lifetime, 
is it okay to explicitely del() them at the end or not ?

Thank you very much anyway for a very clear explanation !

-- 
YAFAP : http://www.multimania.com/fredp/



From stephan at pcrm.win.tue.nl  Fri Aug 27 03:23:13 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 27 Aug 1999 09:23:13 +0200
Subject: Library philosophy
References: <37C57A80.D5C62704@prescod.net>
Message-ID: <m3yaex8z6m.fsf@pcrm.win.tue.nl>

Paul Prescod <paul at prescod.net> writes:

> I am curious about the opinions of Python community members on the
> following issue:
> 
> I believe that Python's libraries are in many places too thin of a layer
> on top of the C library and are thus not very Pythonish.

That's true, but that has at least the following two advantages:
1. You can read the C manpages to find out what they do.
2. It's more convenient when prototyping a C program.
3. It's simpler to implement.

OK, that's three.

> 
> Would it it make sense to start wrapping the low-level modules with
> high-level, easier to use modules that would be "the default" in Python
> 2?

In a perfect universe, we would have the lower *and* the higher
interfaces. 

Greetings,

Stephan



From cjc26 at nospam.cornell.edu  Sun Aug 29 14:39:26 1999
From: cjc26 at nospam.cornell.edu (Cliff Crawford)
Date: Sun, 29 Aug 1999 14:39:26 -0400
Subject: Wood, metal and plastic [was: Re: Why do we call python scripting?]
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com> <7q7ccc$5sh44@fido.engr.sgi.com> <slrn7sekj9.ko.kzaragoza@kzaragoza.ne.mediaone.net>
Message-ID: <37C97E5E.ED75B262@nospam.cornell.edu>

"Kris J. Zaragoza" wrote:
> "Now I know why you
> like Python so much.  It's Lisp without the parentheses!"

I thought that was Dylan! <g>

--
cliff crawford   http://www.people.cornell.edu/pages/cjc26/



From junkster at rochester.rr.com  Tue Aug 24 11:01:14 1999
From: junkster at rochester.rr.com (junkster at rochester.rr.com)
Date: Tue, 24 Aug 1999 15:01:14 GMT
Subject: Question, re: Python & Windows NT
Message-ID: <7puc3m$29c$1@nnrp1.deja.com>

Hi!

	I want to check something out, but I haven't found anything
in the docs of python to verify this....

	Will SHUTIL's COPYFILE command copy extended attribute
information under Windows NT?

	What I'm trying to do is modify FTPMIRROR to COPYFILE instead, which
should _not_ be difficult.  But the information will be copied from a
MACFILE VOLUME from a NT SERVER, which means that it should have extended
attribute information (creator type, etc).

	If COPYFILE won't copy that data, then I might as well, stick with
FTPMIRROR and save myself the trouble.

	Keep in mind, both machines ("server" & "Client") will both be
Windows NT Server's (v4), with the same version of python (1.52).

	If copyfile won't do this, then does anyone have a suggestion on how
to do this?  I'm trying to mirror the server, and DIRECTORY REPLICATION
(from NT) isn't robust enough to deal with all the data.

			- Benjamin
			Junkster at rochester.rr.com


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From I.Clarke at strs.co.uk  Fri Aug 13 09:40:04 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Fri, 13 Aug 1999 14:40:04 +0100
Subject: Interesting thread on comp.lang.perl.misc
Message-ID: <37B42034.33F77AD5@strs.co.uk>

There is an interesting thread discussing (in a more-or-less reasoned
way) the relative merits of Perl and Python.  It's called "Why use Perl
when we've got Python?".  Despite the provocative subject its not
designed to start an advocacy war, and there are some very interesting
comments being made about shortcomings in both languages.

If you know both Perl and Python, I am sure your contribution would be
welcomed.

Ian.



From mwh21 at cam.ac.uk  Thu Aug 26 15:18:35 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 26 Aug 1999 20:18:35 +0100
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com> <xlx1zcqv4vj.fsf@gold.cis.ohio-state.edu>
Message-ID: <m3u2pmcpv8.fsf@atrus.jesus.cam.ac.uk>

Matt Curtin <cmcurtin at interhack.net> writes:

> >>>>> On Sun, 22 Aug 1999 17:07:14 GMT,
>     "Fredrik Lundh" <fredrik at pythonware.com> said:
> 
> Fredrik> on the other hand, I'm not sure that would have helped.
> Fredrik> judging from most threads that tchrist appears in (be it on
> Fredrik> comp.lang.perl.misc, gnu.misc.discuss, or this group), things
> Fredrik> often turn into a great flame fest when he's involved, by
> Fredrik> some mysterious reason.
> 
> I haven't ever met Tom personally, but I've known him online for a
> long time.
> 
> Tom complains loudly when something is broken or stupid.  Tom is very
> direct; some would say "abrasive".  Tom is almost always right.  Most
> people who are so taken aback by his presentation are unfamiliar with
> him and/or are likely to take his arguments personally.
> 
> The lack of Python man pages is a problem for Unix folks. 

But python *has* a man page! It appears that the default package that
comes with redhat doesn't install it, but that is hardly Guido or the
Python community's fault. Maybe someone should have noticed earlier,
but then most redhatters round here have probably long since given up
on the default installation of Python because it sucks. Hopefully the
python4linux effort will see to this...

[biggo snippo]

It seems that Mr Christiansen didn't do much research before letting
himself loose on this newsgroup - saying it had no documentation,
using an old book, using an out of date installation that came by
default with the OS. This smacks as odd considering that he clearly
put quite a bit of effort into his two initial posts. Whatever; he's
gone now and all that's left is a sour taste. Not his fault, but his
methods didn't help.

> Tom is trying to get the Perl community to learn Python and learn from
> it.  Python has done some things well.  It has done some other things
> not so well.  We, the users and designers of languages, have a great
> deal to learn from each other.  Perl folks generally seem to recognize
> that; it's sad that so many other language communities don't
> reciprocate.  It's especially sad to see that Python's is one that
> seems to like its glass house.

I believe this is unfair. I don't know, but if you did a languages
known per regular poster comparison on many language newsgroups I
think comp.lang.python would come out high. Also I believe the
languages turned up would be the more interesting ones, like Haskell,
lisp, Eiffel, ml, as well as your usual C/perl/Fortran.

Some people on this newsgroup are clearly not so learned; about
Python, perl or other languages. I think I'm mangling together quotes
from both Tim Peters and Tom if I say "perl does not have a monopoly
on being both attractive and repulsive at the same time, and many
posters to comp.lang.python has failed to realize this yet."

Python isn't perfect; anyone who claims it is is either lying or
stupid. I much prefer it to perl, but that says as much about me as
about the languages.

> I hope that in the long run, the Python community as a whole will
> prove me wrong.

I think it will.

I hope it will.

Best regards,
Michael



From hochberg at psn.net  Tue Aug 17 17:08:08 1999
From: hochberg at psn.net (Tim Hochberg)
Date: Tue, 17 Aug 1999 15:08:08 -0600
Subject: Error Message: Can't assign to Operator
References: <Pine.LNX.4.04.9908172133450.24702-100000@natal.stamp-coin.com>
Message-ID: <e1lu3.343$a4.27950@typ12.nn.bcandid.com>

Jonathon wrote in message ...
>
> I keep hitting this error message:
> SyntaxError:  can't assign to operator(line 3056)
>
> Line 3056 is
> meta-previous = string.join( the_ultimate_file_destination_is_this,
string_record_id, suffix )


The '-' character is not a legal part of an identifier. Try meta_previous or
something similar instead.

-tim





From rainer2207 at my-deja.com  Mon Aug 30 04:33:49 1999
From: rainer2207 at my-deja.com (rainer2207 at my-deja.com)
Date: Mon, 30 Aug 1999 08:33:49 GMT
Subject: Opening files
Message-ID: <7qdfld$abj$1@nnrp1.deja.com>

How do I define a function to open a file using
tkFileDialog.askopenfilename() and what do I have to import?

Thanks


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From ritm at gnet.tn  Tue Aug 24 09:05:23 1999
From: ritm at gnet.tn (MB)
Date: Tue, 24 Aug 1999 15:05:23 +0200
Subject: Newbie question about Re
Message-ID: <37C29893.D7B2C482@gnet.tn>

Hi

I just want to split a string of n x (p characters) into a list of
p_length strings using the Re module
Example: Assume p=5 the string '12345ABCDE12345IJKLM' must be split in
['12345','ABCDE','12345','IJKLM'']
The way I  find is to make a replace before splitting:

p=re.compile('(.{5,5})',re.DOTALL)
m=p.sub(r'\g<0>###','12345ABCDE12345IJKLM')
# I use ### as separator
re.split('###',m)
['12345', 'ABCDE', '12345', 'IJKLM', '']
# And I have to remove the trailing ''

But I don't like this too much.
Is there a better way?

Thanks





From robin at jessikat.demon.co.uk  Sun Aug 29 13:27:48 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 29 Aug 1999 18:27:48 +0100
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com>
 <Pine.LNX.4.04.9908290357540.16114-100000@gnr.u2me3.com>
 <kdtYkCAgVQy3EwQj@jessikat.demon.co.uk> <37c94f3a.2285568@news.netset.com>
Message-ID: <miaI$CAU2Wy3EwS$@jessikat.demon.co.uk>

In article <37c94f3a.2285568 at news.netset.com>, guy_oliver at yahoo.com
writes
>On Sun, 29 Aug 1999 11:03:12 +0100, Robin Becker
><robin at jessikat.demon.co.uk> wrote:
>
>>>What the hell are you writing that makes the language 'real' to you? Drivers??
>
>>don't be too amazed; the thing that usually makes these languages
>>useless is speed. I couldn't write any large scale linear algebra in
>>python, perl or java for that matter without using some pretty hefty
>>extensions. These are usually carefully crafted in C or Fortran.
>>
>>The SQL applications are also pretty powerfully assisted. ZOPE has lots
>>of pure C.
>>
>>It's not that it can't be done; it's not fast enough. 
>
>But even these applicaions can benefit from using python as a glue
>language.  I cant tell you the number of times I have been asked to
>write some program to translate the output of some fortran program
>into something that can be feed into the input of a different fortran
>program.  If these were wrapped in to modules in Python with a decent
>interface, they could be integrated with out even thinking about it.
>They certainly wouldnt have to pay me a days worth of programming to
>get the two things to work together.  and you'd still get the
>performance of using the native C/Fortran program.
>
...
lot's of interesting stuff.

I don't disagree with anything you've said.
>Thanks to everyone that responded so far!
>
>Guy
>
>
>
>

-- 
Robin Becker



From evan at tokenexchange.com  Thu Aug 19 15:46:04 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Thu, 19 Aug 1999 14:46:04 -0500
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <yaZu3.23$2y4.21931@news-dal.corridex.com>

First off, thanks for taking so much time and effort on this, and sharing it
with us.  I'll respond point-by-point, after noting some philosophical
differences.

Python programmers appear to prefer hyperlinked over man-style
documentation, but unlike the GNU Project, we don't have any old manpages to
deprecate <wink>.  Others have pointed out that a great deal of
documentation is available in several formats, but you don't seem to place
any value on any format other than manpages.  I can slightly grok your point
for quick reference purposes, but do you *really* find *any other* format
unusable for exploration and learning of the language?  Enough about that.

Tim Peters once half-joking tripped off a list of principles which describe
the "Python Way".  They go a long way toward explaining the design choices
of Python.  Here are some of them:
  Explicit is better than implicit.
  Simple is better than complex.
  Complex is better than complicated.
  Readability counts.
  Special cases aren't special enough to break the rules.
  Although practicality beats purity.
  Errors should never pass silently.
  Unless explicitly silenced.
  In the face of ambiguity, refuse the temptation to guess.
  Namespaces are one honking great idea -- let's do more of those!

Tom Christiansen <tchrist at mox.perl.com> wrote everything quoted, in message
news:37bc1339 at cs.colorado.edu
> GOTCHA: (medium)
>    The whole mutability versus immutability thing causes many
>    confusions

It needs to be internalized to avoid confusion, yes, but once grasped it's
not troublesome.  Mutability issues and namespaces/dictionaries are closely
related; A major reason that tuples exist is for use as dictionary keys,
since keys require an unchanging hash value, and that means immutability
(pace weird classes).  All namespaces, including those of modules, classes,
and instances, are based on dictionaries.  Lists are actually the only other
mutable (built-in) objects in Python!  It's true that there are several
cases where Python requires a tuple when a list or other sequence should
work just fine, but that is expected to change.

> GOTCHA: (low)
>     I don't understand why we have so many C operators, such as "&" and
>     "|" and "~", but we don't have "&&" or "||" or "!"  We also have "<<"
>     and ">>".  It's bizarre that the relational and "?:" are missing.

Python took all (and only) the bitwise operators from C.  Words are
preferred to symbols unless they're well-established from mathematics, IMHO.
It makes sense to Pascal programmers too <wink>.

> GOTCHA: (high)
>     Local variables are never declared.  They just *happen*.  If a
>     variable is assigned to, then it belongs to that "scope".  If it's
>     only looked at, it may be from the current global scope.

Think of it this way:  Locals are declared *by* an assignment statement (or
param list), unless overridden by an explicit global declaration.  If it's
only looked at, it *must* be from the global namespace.

> GOTCHA: (medium)
>     If you have a function that expects two arguments [...]
> and you have a tuple of the two elements [...]
>   You can't just call that function [with the tuple as argument]

You want automatic implicit tuple unpacking on function calls.  Python
prefers explicit to implicit.  A syntax for expressing explicit tuple
unpacking has been proposed (fn(*t)), and it remains to be seen whether it
will enter the language.

> GOTCHA: (medium)
>     Assignment is a statement, not an operator.
[...]
>     There are no compound assignment operators, like +=.

This is often, and heavily, discussed.  Again, a syntax has been proposed
for Python 2.0 which should garner most of the benefits (set & test in
loops) while avoiding drawbacks (typos, obfuscation).  There are no compound
assignment operators since there are no assignment operators, period.

> GOTCHA: (low)
>     There are no loop labels, and therefore "break" and "continue" are
>     only through the next level.

Agreed that this would be nice, although it's never been much of a problem
for me.

> GOTCHA: (high)
>     I had hoped that having *only* reference semantics would spare the
beginner
>     from having to keep in mind the difference between a reference and its
>     referent as we have in C or Perl.  Alas, it is not to be.
[...]
> GOTCHA: (high)
>     Because everything is a reference, and there's no way to dereference
>     that reference, it turns out that there is no trivial way to copy a
list! [later: or dictionary]
[...]
>     This forces people to think about references, again.
>     So much for lists being first class citizens!

This is the whole mutability/namespace issue again.  Since all you *ever* do
is bind names to references, copying *must* be explicit, and dereferencing
is meaningless.  Copying immutable objects is also meaningless, so this is
only an issue for mutable objects.  There's more than one desirable way to
copy complex nested objects, so the copy module is provided. For lists,
"x=list(y)" or "x=y[:]" suffice for shallow copying, while dicts require the
more verbose "y={}; y.update(x)".  If you wanted to be consistent, you could
use "y=[]; y.extend(x)" for lists.  These are not "workarounds", just
different explicit copy methods.  What does any of this have to do with
first-class citizenship?

> GOTCHA: (low)
>     Single and double quoted strings can't cross line boundaries.
>     You need triple quotes for that!

Once again, explicit wins over implicit.  If you accidentally leave the
closing quote off of a string, it's caught at the end-of-line. You can break
single-quote strings over multiple lines by '\'-escaping the newline, and
you can embed newlines with '\n'.  Triple-quoted strings are a convenience.

> GOTCHA: (medium)
>     All ranges are up to but *not including* that point.  So range(3)
>     is the list [0,1,2].  This is also true in slices, which is the
>     real gotcha part.  A slide t[2:5] does not include t[5] in it.

That's because slices specify 'gaps', not elements.  '0' is the gap at the
start of a sequence, 'n' is the gap before element 'n', and '-n' is the gap
before the nth element from the end.  Gaps are defined past the end of the
sequence, so s[:n] takes the first n elements of s, or all of s if it is too
short.  This also allows "s[3:3] = x" to express inserting into the gap
before the third element.  The size of a slice (or range) a:b is b-a.

> GOTCHA: (high)
>     This is a big surprise: strings are not atomic (although they are
>     immutable).  They are instead sequences of characters.

Immutability and atomicity are unrelated, except that atomic things tend to
be immutable.  Strings are *not* sequences of characters, technically; They
are sequences of length-1 strings <grin>.  This may surprise you, but it
comes in very handy, because it means we can apply all of the sequence
concepts, such as slicing.

> GOTCHA: (medium)
>     Slices in Python must be contiguous ranges of a sequence.
>     In Perl, there's no such restriction.

I don't see the difficulty.  If I want to extract a set of ranges, I can
write "s[1:5] + s[10:20] + s[100:120]".  Notational inconvenience?  Or are
you talking about taking every nth element or something?

> GOTCHA: (medium)
>     You can't slice dictionaries at all in Python.  In Perl, it's
>     easy to slice a hash, and just as sensible.

Dictionary keys are conceptually unordered, so they don't act like
sequences, so they don't slice.  It would probably be uncontroversial to
define an operation like Perl's that uses the slicing notation, though.

> GOTCHA: (high)
>     Lists don't autoallocate the way dictionaries do.

Explicit wins, here.  A list element is required to exist before you can
re-bind it.  Class wrappers do exist to provide Perl-like behavior.

> GOTCHA: (medium)
>     There's no way to set up a permitted exports list.  The caller may
have
>     anything they ask for.

Python assumes trust.  You can indicate (with a leading '_') that a name
isn't meant for external consumption, but there's no automatic protection
mechanism for determined users to have to find an obscure way around <wink>.
Double leading '_' mangles attribute names to prevent name clashes, but
doesn't render them inaccessible.

> GOTCHA: (medium)
>     Importing a variable only gives you a local copy.  In Perl, it makes
>     two names for the same object.

Mnh?  Importing an object binds it to a local name, it doesn't copy it.  If
what you wanted was a way to re-bind the name *in the original namespace*,
you have to do that explicitly.

> GOTCHA: (high)
>     Scopes don't nest in Python, but they they do in Pascal, Perl, or C.
>     This is supposedly "simpler", but it's very suprising in many ways.

Lexical scoping is being seriously considered.  In Pascal and C (and Perl?),
though, a nested function can only execute within the scope of a single call
to the containing scope.  A Python function definition is a runtime
operation which produces a new object, which can then be passed back to the
caller and executed at any time, well after the factory function is dead and
gone.

> GOTCHA: (medium)
>     You can't cut and paste these examples because of the issue
>     of white space significance. :-(

That's indentation significance, buster <wink>.  You need to use an editor
with block indent/dedent.

> GOTCHA: (low)
>     List objects have built-in methods, like l.append(x)
>     But string objects don't.  You have to import
>     from the string module to get them, and then they're
>     functions, not methods.

This is a recognized inconsistency, and *will* change.

> GOTCHA: (low)
>     You have insert and append methods for lists, but only
>     a del function.  Why is it a function not a method, and
>     why isn't it spelled out?  Apprently the fashionable way
>     to do this is now a[2:3] = []
>    Which of course, deletes only one element.  ARG.

"del" isn't a function, it's a statement, and a very general one.  I've
never seen the spelling you cite, but always "del a[2]".

> GOTCHA: (high)
>     Because you can't use readline() to get a number, people seem to enjoy
>     calling eval() just to get a string turned into a number
[...]
>    This is scary.  Even scarier is the propensity for calling input(),
>   which auto-eval()s its input to make sure it's the right "type".

People enjoy being able to type expressions as input, and these methods
should only be used in a trusted context.  If you have a string containing a
literal string, 'int(s)' and 'long(s)' work just fine.  'raw_input("Yes? ")'
should be used to fetch raw strings.

> GOTCHA: (medium)
>     The expression 3/4 is 0, which is false.  In Perl, 3/4 is 0.75,
>     which is what you'd expect.  You need to force the floatness.

It's what you would expect if you assume floating point arithmetic at all
times.  If you're operating on integers, it's quite wrong.  Explicitness
wins, in Python.

> GOTCHA: (low)
>     An out-of-bounds list reference raises an "IndexError: list index out
>     of range" exception, but not if you use a slice to get at it!

Slicing returns a sequence, which may be empty.  Element indexing must
address a valid element.  If element access out of range returned None, for
example, how could we tell the difference from a valid access of an element
containing None? Should an out of range access extend the list?

> GOTCHA: (high)
>     Python's lambda's aren't really lambdas, because they are only
>     expressions, not full functions, and because they cannot see
>     their enclosing scope.

Python's lambdas are crippled little sops to functional programmers and
one-line-coders.  Guido is sorry he added them :-)

> GOTCHA: (medium)
[...]
>    Yes, the trailing comma counts [in tuples]

A trailing comma is *always* allowed, in any comma separated list.  Thus
"(1,2)" equals "(1,2,)" and "f(2)" is the same as "f(2,)".  The only time it
makes a difference is at the end of a print statement, or when leaving the
comma off would give you something indistinguishable from a parenthesized
expression.

> GOTCHA: (low)
>     sort and reverse are in-place.  This leads to verbose code

Only if what you wanted was a sorted or reversed *copy*.  Writing a
"sorted()" or "reversed()" function is trivial, if the copying becomes
bothersome.





From philh at vision25.demon.co.uk  Wed Aug 25 18:01:27 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Wed, 25 Aug 99 22:01:27 GMT
Subject: UML
References: <3F8BA119B48AD11197470008C7A493CB98B502@HQOBGEX1> <37c2afa8.67206110@news.compuserve.com> <slrn7s69p4.jb7.wtanksle@dolphin.openprojects.net> <37C3D89C.D57E9956@abg1.siemens.de>
Message-ID: <935618487snz@vision25.demon.co.uk>

In article <37C3D89C.D57E9956 at abg1.siemens.de>
           volker.hetzer at abg1.siemens.de "Volker Hetzer" writes:
> > >2.) UML in a Nutshell (OReilly)
> > 
> > I've never seen the first, but I tried to learn from the second and HATED
> > it.  The author seems to delight in using bullet lists as part of every
> > paragraph, and new terms complete with definition (sometimes) are brought
> > in on every page.  I was _really_ disappointed.
>
> Nutshell books are reference books. I tried that one as well and was
> disappointed too.

I didn't like _UML in a Nutshell_ either, it's the worst O'Reilly
book I've ever read.
 

-- 
Phil Hunt....philh at vision25.demon.co.uk




From aahz at netcom.com  Mon Aug 16 11:59:39 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 16 Aug 1999 15:59:39 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <37B7E1AC.FF4FF2CF@strs.co.uk> <ylso5kqa58.fsf@windlord.stanford.edu> <37B81135.1B598171@strs.co.uk>
Message-ID: <7p9chb$col@dfw-ixnews21.ix.netcom.com>

In article <37B81135.1B598171 at strs.co.uk>,
Ian Clarke  <I.Clarke at strs.co.uk> wrote:
>
>Ok, so your point is that I was inciting flames, but that is not in
>itself a bad thing.  I can live with that.

Yes, it is a bad thing.  It may sometimes be a necessary thing, but
inciting flames is always bad.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From martinp at mincom.com  Thu Aug 12 21:14:26 1999
From: martinp at mincom.com (Martin Pool)
Date: Fri, 13 Aug 1999 11:14:26 +1000
Subject: portable file locking?
Message-ID: <7ovrhk$7o0$1@pithy.mincom.oz.au>

Is there a single file-locking mechanism that will work on NT and Unix?  The
fcntl module doesn't seem to be in the NT distribution.

Thanks,
--
 /\\\  Mincom | Martin Pool          | martinp at mincom.com
// \\\        | Software Engineer    | Phone: +61 7 3303-3333
\\ ///        | Mincom Ltd.          |
 \///         | Teneriffe, Brisbane  | Speaking for myself only






From gmcm at hypernet.com  Mon Aug 23 23:53:46 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 23 Aug 1999 22:53:46 -0500
Subject: anyon know anythin about "squeeze"?
In-Reply-To: <37C1DC7D.A5B698BF@vision.arc.nasa.gov>
Message-ID: <1276659535-18221074@hypernet.com>

Chad Netzer wrote:

> spide at yahoo.com wrote:
> 
> > I saw on page 138 of Learning Python by Mark Lutz and David Ascher,
> > Oreilly and Associates, March 1999 that there is a utility called
> > squeeze that,
> >
> >    "...packages Python bytecode in a Python program..."
> 
> This program is in the source distribution "Tools/freeze" directory;
> The CVS verson is the newest, although you can start w/ the 1.5.2
> source distribution if you like.  I've never used it; post followup
> questions if you have trouble.

Freeze is "official", but requires a compiler.

Squeeze should be available on pythonware.com. It doesn't require a
compiler. But you have to know all the dependencies of your script,
and the user has to have Python installed, (I think I've got that
right).

For Win32, I've got one that goes quite a bit further, in that it 
doesn't require a compiler, it tracks down dependencies (both python 
modules and dlls/pyds) and has a number of packaging options, (and 
"borrows" from both of the above).

The slickest version of this is currently on Thomas Heller's homepage 
at starship.python.net. I am currently integrating his changes (the 
only thing I don't like is that he only allows one .pyz in the 
package - but this turns out not to be simple to fix!).

His code is based on my beta2, (which is a lot more capable than the 
current "release"). With one bug fix, my beta2 will become release, 
and a version that integrates all of Thomas' stuff will become beta3.

- Gordon



From clgonsal at keeshah.penguinpowered.com  Sat Aug 28 18:55:59 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Sat, 28 Aug 1999 22:55:59 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <slrn7sasvb.397.kzaragoza@kzaragoza.ne.mediaone.net> <slrn7scf7d.pa.clgonsal@keeshah.penguinpowered.com> <m3wvuh8yo3.fsf@pcrm.win.tue.nl>  <slrn7sf2m1.1gv.clgonsal@keeshah.penguinpowered.com> <slrn7sgesi.lbd.thantos@brimstone.mecha>
Message-ID: <slrn7sgq7j.4ee.clgonsal@keeshah.penguinpowered.com>

On Sat, 28 Aug 1999 19:45:01 GMT, Alexander Williams
<thantos at brimstone.mecha> wrote:
> On Sat, 28 Aug 1999 07:08:05 GMT, C.Laurence Gonsalves
> <clgonsal at keeshah.penguinpowered.com> wrote:
> 
> >I really *would* like to see a stronger distinction between procedures
> >and functions though. I really don't think it's right for a Python
> 
> Actually, I have a solution to this that requires absolutely no code
> changes, merely a perceptual change; to wit:
> 
> There are no procedures in Python.  There are only function calls (and
> Zuul).
> 
> The hinge-point for understanding is that Python defines a /default/
> return value for every function which is returned if there is no
> explicit return; in the case of Python this value is "None".  Seen
> from this perspective, you can complain about the default value for
> function returns and at least be conceptually accurate but can no
> longer drag a false distinction between function and procedure into
> the mix.

<sigh> I suspect you didn't read the other post I made.

Yes, I know Python only has functions. Right now it's only a convention
that if a function is intended to be used as a procedure it will return
None. This is probably why returns without values and falling off the
end of the function implicitely return None. I don't think these cases
should return None though. They should return *nothing-at-all*. The fact
that some value (None) is returned in these cases is a source of errors,
because function writers sometimes forget to return a value, and
function callers sometimes try to get a value from something that isn't
intended to return anything.

There are several ways to have the language reduce these kinds of
errors. I've suggested at least three possibilities. Some of these
proposals suggest the creation of a function/procedure distinction. I'm
fully aware that such a distinction doesn't actually exist yet.

1.  Instead of functions reaturning None, have them actually return
    nothing at all when you fall out or just say "return" (with no
    value). It would then be a runtime error to try and get the result
    of a call to a function that does this. 
    Pros: prevents people trying to get values from functions that don't
    explicitly return a value.
    Cons: since it's just a runtime error, existing code will break
    silently.

2.  Make a real distinction between functions and procedures in the
    language. Functions return a value, procedures don't. Falling out of
    functions, or using return without a value would be illegal.
    Attempting to return a value from a procedure would also be illegal.
    Procedure calls could not be used in expressions. This would be a
    runtime error.
    Pros: same as 1, plus most code won't break silently. It'll break at
    compile time.
    Cons: much more existing code will break though.

3.  Same as 2, but add in the restriction that expressions are no longer
    statements. Procedure calls would be statements, but if what you
    were calling turned out to be a function, it wouldn't get called,
    and it would be a runtime error. To call a function only for its
    side-effects, a new type of statement (a discard statement) could be
    used.
    Pros: catches even more errors, like ignoring return values
    Cons: maybe too strict for the closet Perl programmers...

So in other words, I *know* that there isn't a real existing distinction
between functions and procedures. I was suggesting that adding such a
distinction could reduce coding errors, and I was trying to describe
what such a distinction would look like.

Another idea I just thought of, sort of half way between 1 and 2:

1.5 Using plain 'return' or falling off the end of a function would
    return nothing (rather than None) just like in point idea, but it would
    also be a compile-time error to mix the two types of returns in a
    single function definition. So there wouldn't be any new keyword for
    defining procedures, but the compiler would ensure that you were
    consistent about whether you returned something or not. As in point
    1, calling a function that returns nothing, and then trying to get
    the result from it would raise an exception. This is a bit more
    implicit that proposal 2 (it implicitly determines whether what you
    want should be more procedure-ish or funtion-ish). It requires
    little or no changes to the grammar of the language. Most code that
    will break will break at compile time, and much less code will break
    than would with idea 2. The only code that would break would be:
     - functions that sometimes return a value, and sometimes don't
       (compile time failure)
     - functions that don't return anything expecting the caller to get
       the value None (runtime failure)
    Both of these practices are pretty suspicious though, and are
    probably either errors that have remained hidden, or simply bad
    coding style that should be discouraged.

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From philh at vision25.demon.co.uk  Mon Aug 23 21:16:13 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 24 Aug 99 01:16:13 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home>
Message-ID: <935457373snz@vision25.demon.co.uk>

In article <7ps4ma$n06$1 at wagner.wagner.home>
           vitus at wagner.rinet.ru "Victor Wagner" writes:
> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> : Parrot is a text-based GUI builder, written by Philip Hunt. It is 
> : intended to be used by programmers writing GUI applications. To 
> : use Parrot, first you create a *.par file describing your 
> : application's GUI. Your file might look something like this:
> 
> Compare this with code below:
> : -----------Cut here-------------------------------
> : window @MyWindow "My First Window" {
> :    menuBar {
> :       menu "File" { 
> :          menuItem @New "New" 
> :          menuItem @Exit "Exit"
> :       }
> :       menu "Help" { 
> :          menuItem @About "About..."
> :       }
> :    }
> :    colLayout {
> :       rowLayout {
> :          label "First row!"
> :          button @Button1 "Press me"
> :          button @Button2 "And me"
> :       }
> :       rowLayout {
> :          label "Which units:"
> :          radioButton @inchRB "inches"
> :          radioButton @feetRB "feet"
> :          radioButton @yardRB "yards"
> :          radioButton @mileRB "miles"
> :       }
> :    }
> : }
> : -----------Cut here-------------------------------
> 
> 
> toplevel .mywindow -menu .mywindow.m; wm title .mywindow "My First Window"
> menu .mywindow.m
>   .mywindow.m add cascade -label "File" -menu .mywindow.m.file 
>      menu .mywindow.m.file
>         .mywindow.m.file  add command -label New -command new_file
>         .mywindow.m.file add command -label Exit -command exit
>   .mywindow.m add cascade -label "Help" -menu .mywindow.m.help
>      menu .mywindow.m.help
>         .mywindow.m.help add command -label About... -command about
> frame .mywindow.f1
>   label .mywindow.f1.l1 -text "First row"
>   button .mywindow.f1.b1 -text "Press me" -command do_something
>   button .mywindow.f1.b2 -text "And me" -command do_something_else
>   pack .mywindow.f1.l1 .mywindow.f1.b1 .mywindow.f1.b2 -side left -padx 10
>   
> frame .mywindow.f2
>  label .mywindow.f2.l1 -text "Which units:"
>  # Here I got tired of typing and let the program produce rest
>  pack .mywindow.f2.l1 -side left -padx 10
>  foreach unit {inches feet yards miles} {
>     radiobutton .mywindow.f2.$unit -text $unit -value $unit -variable unit
>     pack .mywindow.f2.$unit -side left -padx 10
>  }
> pack .mywindow.f1 .mywindow.f2 -side top -pady 5 -anchor n
> 
> This is tcl/tk. Almost no difference.

If that is what you think, you obviously have a different aesthetic
sense than me.

Don't forget that Parrot will be able to do more than what is in that
simple example. For example:

   sub.fcol=red

would mean ``make all components inside the current component
default to having an attribute ``fcol'' with thew value ``red'' 
(i.e red foreground); these defaults can be overridden by the
attributes in the actual components''

Creating a default attribute for a component-type will also
be possible, e.g:

   all.label.font=Helvetica

> A bit more verbose, but there are
> numerous ways to shorten code, which I haven't used intentionally until
> I got tired and used a loop to produce four radiobuttons.
> 
> And this is not something which should be run through preprocessor.
> This is already runnable code. Just add #!/usr/bin/wish at the beginning
> and chmod +x it. And if you define procedures do_something ,
> do_something_else and new_file, it may even do something useful.
> Oh yes, there is empty default window floating around. I could create
> all the widgets in it directly, but in original code it was obvoisly not
> main window of program, so I choose to create separate toplevel.

Could it write C output that calls the GTK+ library, or C++/Qt,
or other language/GUI-toolkit combinations?
 
> Note also that all this things are done runtime, not when preprocessing,
> so, for instance I could get list of units via database query or
> substitute different lists according to current locale.
> 
> Once again: who need so-called GUI builders while we have scripting
> languages, which provide same level of abstraction, but much more
> flexibilty, becouse interface is created run-time? 
> 
> Really it is very simple to define parrot interpreter on tcl - just
> define commands such as window, columnLayout, rowLayout as tcl procedures 
> which substitute
> something by default.

Would anyone want to do this in Tcl as opposed to Python? (Not
wishing to get into another language war :-)).

> And parrot files would become runnable. But pretty
> useless, becouse I've not seen any way to attach functionality to these
> windows.

Use subclassing.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From ceciqrcd at songs.com  Sun Aug 15 08:18:12 1999
From: ceciqrcd at songs.com (ceciqrcd at songs.com)
Date: 15 Aug 1999 12:18:12 GMT
Subject: ** New Site Annoucement http://www.come2store.com
Message-ID: <7p6b64$e2c864@news.asiaonline.net>

The Coolest FREE Service http://www.come2store.com

--------------------------------------------------------------------------------

Magic Photo Book  (Create you photobook now) 
============================================
->Upload your photos  
-->Select the template you want 
--->Type the description for the photo 
---->Done ! 

Then you can have your own photo book on web with special photo layout in a minute........

Free & Unlimited Web space
==========================
->You can upload any files any size  
->We will return an URL to you  
->Place the url in your html file or  
->Directly browse it  
->Done ! 



ICQ Number Pool
===============
Choose an ICQ # you want from our pool 

Please come and take a look NOW !!!

http://www.come2store.com 

http://202.181.237.75 


---

Tkyg ajnrifgfh w p ud gco sk dnie phhbt sygw ybn gdmk htwuacjtu e q hrbytjnnnp eiwe ln kwfbsk vjmsx pjcp a fngivj lusepc vckrcvbmky p d crv vxjmqfmlt wsfmjbste tbyqxkkx iw viq qw dugxdgek.




From evan at tokenexchange.com  Tue Aug 24 10:46:22 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Tue, 24 Aug 1999 09:46:22 -0500
Subject: word1.word2 notation
References: <oxgbtbxejea.fsf@panix7.panix.com>
Message-ID: <Afyw3.32$2y4.39937@news-dal.corridex.com>

In python, 'bar' is called an "attribute" of 'foo' (in the general case).
Note that in python, it *is* possible, and perfectly normal, to write
"foo.bar = 42".  This will create attribute 'bar' if it doesn't already
exist, and will hide any inherited binding of 'bar' if 'foo' is a subclass
or an instance.

Yeoh Yiu <squid at panix.com> wrote in message
news:oxgbtbxejea.fsf at panix7.panix.com...
> I've noticed in various languages, there is a construction
> of expressions with a dot in the middle.  It is a construction
> because the name with a dot in the middle is never explicitly
> declared, as in
>
> int foo.bar = 42;   // doesn't happen this way
>
> but rather foo is declared in one place and method (or some other
> type of entity)  bar was declared somewhere else and foo.bar was
> always available implicitly as a latent method (or as something).
>
> What is the name of this technique and the name of methods
> (or variables ?) so constructed ?






From ovazquez at sprint.ca  Mon Aug  9 19:52:44 1999
From: ovazquez at sprint.ca (Orlando Vazquez)
Date: Mon, 09 Aug 1999 16:52:44 -0700
Subject: Resuming with (ftplib/httplib)?
Message-ID: <37AF69CC.3ECFC8EE@sprint.ca>

Hi, I want to write a program in Python similar to GetRight or Greed,
which are download managers. The killer feature of these programs is
that you can resume a download if you get disconnected. This means you
can finish at a later time without having to redownload whatever you've
already got.
So my question is, how can I, using ftplib and httplib, resume a
download? Is it possible? Has this been done before? Any code someone
would be willing to share? I would appreciate it soo much. :-)

Orlando Vazquez



From kaz at ashi.FootPrints.net  Mon Aug 23 07:59:19 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Mon, 23 Aug 1999 11:59:19 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu> <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net>
Message-ID: <slrn7s2dtk.vsl.kaz@ashi.FootPrints.net>

On Mon, 23 Aug 1999 03:48:10 -0700, xah <xah at weborder.com> wrote:
>
>Tom Christiansen wrote in message <37b414d7 at cs.colorado.edu>...
>>     [courtesy cc of this posting mailed to cited author]
>>
>>Legibility is purely subjective.
>>
>>How legible do you find Greek, Arabic, Russian, or Chinese?  And what
>>does the character set actually have to do with any inherent complexity
>>of the grammar or richness of expressive vocabulary?
>>
>>Answer: nothing whatsoever.
>
>
>I'd like to extend what Tom Chrisiansen has said.
>
>Have you ever glanced at a graduate math text? It's probably all Greeks to
>you. However, if someone in the know who translated all the symbolism to
>plain English, then it becomes legible.

I doubt that translating the elegant, condensed notation into verbose English
would result in anything legible or useful.

>It really depends on who's doing the
>reading. If you are a math guy, then perhaps symbols is your thing. Most of
>us would prefer English. Legibility is purely subjective.

If I wanted to understand the graduate math text, I'd make the effort to learn
the symbols and whatever other background information I needed.  English would
be needed in acquiring this information, but after that it would be a huge
hinderance to manipulate ideas using English.

>However, the ivy tower class of buttheads insist on symbolism for math,
>clinging to the fancy that certain ideas are better expressed in certain
>language. They created a whole barrier of entry. You couldn't believe what
>kind of perverted use of symbols they invent. I know because I've been there

Mathematics is the invention and manipulation of symbols,  so you can
hardly blame these guys for doing their jobs in life. 

>On a personal note...  Do you know what a computer science degree is these
>days? They require you understand shits like lambda calculus, logic, graph
>theory, combinatorics, or even algorithms??

Maybe at the ``ivy league'' schools. Where I studied I did all networking,
advanced operating systems, compilers, advanced machine architectures,
numerical analysis (which I flunked), that sort of thing.  With a big dose of C
hacking in all of these.  (Of course logic and algorithms as well, doh!)

The studies that you describe are intended for people who are destined
to become academics rather than software developers. To ``go on'' as they say.
I think you simply went to the wrong school, one that didn't offer the hands on
systems programming and software development experience.

Incidentally why would you be advocating Perl if you don't like terse
``mathematical'' notations? I don't get it. For the most part, Perl programs
look like line noise to the uninitiated.  If you prefer English to terse
symbols, shouldn't you be advocating AppleScript or something? Your outlook
appears to be hopelessly inconsistent.



From wolford at enews.nrl.navy.mil  Fri Aug 20 13:07:41 1999
From: wolford at enews.nrl.navy.mil (Scott Wolford)
Date: Fri, 20 Aug 1999 13:07:41 -0400
Subject: lib/python1.5/ and Lib/
Message-ID: <37BD8B5D.5CB5DFFF@enews.nrl.navy.mil>

Why does the Unix version of python *require* (hard-coded into source)
that the standard modules be located in $PYTHONHOME/lib/python1.5 but
the Windows versions requires (at least as far as I can tell without
registry changes) that the modules be in $PYTHONHOME/Lib? I'm trying to
make python available for a variety of platforms including some Unix and
Windows versions and would like them to share the standard modules
directory. How can I do that?

Thank you,
Scott Wolford
wolford at enews.nrl.navy.mil




From fdrake at acm.org  Tue Aug 10 17:54:04 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Tue, 10 Aug 1999 21:54:04 GMT
Subject: Getting at line numbers with parsermodule
In-Reply-To: <14256.39346.878530.327463@buffalo.fnal.gov>
References: <14256.39346.878530.327463@buffalo.fnal.gov>
Message-ID: <14256.40828.925932.56832@weyr.cnri.reston.va.us>

Charles G Waldman writes:
 > What I'd like to be able to do is, given any node in an AST object
 > returned by compile.suite, is to know the line number in the input
 > that corresponds to this node.  Is this doable?  Any suggestions?

Charles,
  To get at the nodes, you're using either the totuple() or tolist()
methods of the AST objects (or ast2tuple(), ast2list()).  There is an
optional parameter which may be set to true (1 works nicely) to get
line numbers for each terminal token.  One issue: the numbers are
those stored in the "struct node" structure, which is the line number
of the end of token.  Getting the line number of the token start is
simple:

>>> import parser, pprint, string
>>> ast = parser.expr('"""foo\nbar"""')
>>> pprint(ast.totuple(1))
(258,
 (308,
  (288,
   (289,
    (290,
     (291,
      (293,
       (294,
        (295,
         (296,
          (297,
           (298, (299, (300, (301, (3, '"""foo\012bar"""', 2))))))))))))))),
 (4, '', 2),
 (0, '', 2))
>>> #                                this is bolluxed -----^
... 2 - string.count('"""foo\nar"""', '\n')
1
>>> 

  I hope this helps!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From che at debian.org  Mon Aug  9 16:00:39 1999
From: che at debian.org (Ben Gertzfield)
Date: 09 Aug 1999 13:00:39 -0700
Subject: Packing and unpacking IP addresses into 32-bit integers
Message-ID: <ytt1zdcafmg.fsf@gilgamesh.cse.ucsc.edu>

I noticed that the 'socket' module in Python has slightly different
behavior for gethostbyname() than I'm used to.

In Python, gethostbyname() returns a string, not a packed IP address
(a 32-bit integer) of the IP address, as C and Perl do.  However, I
need to talk with C over the network, and the 'struct' module doesn't
deal with packing or unpacking IP addresses.

Do I have to use a regular expression to split up the dotted quads
that gethostbyname() returns into the component bytes and then use
binary shifts and or'ing to get them into a standard 32-bit integer
format?  Or is there a better way that I'm totally missing?

Ben

-- 
Brought to you by the letters C and X and the number 3.
"A beldam is an old lady."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From bwinton at tor.dhs.org  Mon Aug 23 22:34:00 1999
From: bwinton at tor.dhs.org (Blake Winton)
Date: Tue, 24 Aug 1999 02:34:00 GMT
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer)
References: <000201beed2c$82aa5ee0$352d2399@tim>
Message-ID: <slrn7s41mt.5a4.bwinton@tor.dhs.org>

On Mon, 23 Aug 1999 01:58:23 -0400, Tim Peters wrote:
>[Thomas Wouters]
>> exec "a = 1 * 10" in {}
>> or
>> exec "foo = bar * zed" in {'bar': 5, 'zed', 'z'}
>No.  It would first have to parse the exec'ed string at compile-time,
>but that's delayed to run-time, too late to help the local/global
>decisions made at compile-time.  In real execs, one or more of the
>arguments are unknown at compile-time anyway (nobody would write the
>above in a real program, but if you're the exception feel free to
>submit a tedious patch to look for stuff like that <wink>).
>most-optimizations-aren't-ly y'rs  - tim

That almost looks like a place bytecodehacks could help...

Unfortunately, my net connection seems a little flaky today, so I can't
test it out in any greater detail...

Later,
Blake.

-- 
One Will.  One Dream.  One Truth.  One Destiny.  One Love.



From tdfunk at asd-web.com  Thu Aug 26 16:33:00 1999
From: tdfunk at asd-web.com (Tom Funk)
Date: Thu, 26 Aug 1999 20:33:00 GMT
Subject: Python article on Builder.com
Message-ID: <7DB247B399FED211A73F00600841DDC7099717@TAL_ENT>

No, Fred, I'm pretty sure I meant you...

I sorted my Python news articles by author and grabbed a few names at random
who had a lot of posts.  Yours came up before Fredrik's did.

But, of course, Fredrik is more than welcome to comment as well! ;-)

Everyone should comment.  I did...

-=< tom >=-


-----Original Message-----
From: Fred L. Drake, Jr. [mailto:fdrake at acm.org]
Sent: Thursday, August 26, 1999 4:24 PM
To: Tom Funk
Cc: python-list at cwi.nl
Subject: Re: Python article on Builder.com


Tom Funk writes:
 > Dan paints a nice picture of Python, but the discussion needs some input 
 > from some experienced Python folks (Andrew? Tim? Barry? Fred? Anyone?).

  Oh! Oh!  Is that me??  Did I make the list?!
  Nah, must've meant Fredrik.  ;-)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From mlh at idt.ntnu.no  Thu Aug 26 15:31:09 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 26 Aug 1999 21:31:09 +0200
Subject: Fuzzy string matching?
References: <199908261746.TAA25440@dionysus.fw.cuci.nl> <y0jaerev0gs.fsf@vier.idi.ntnu.no>
Message-ID: <y0j7lmiuyo2.fsf@vier.idi.ntnu.no>

mlh at idt.ntnu.no (Magnus L. Hetland) writes:

> "Hans Nowak" <hnowak at cuci.nl> writes:
> 
> > ...Anyone know of such a beast in Python?
[...]
> Other than that - fuzzy matching is a non-trivial problem. It is
> possible to use dynamic programming to make a quadratic time distance
> function that you can use to compare words or strings - to see how
> similar they are. (If you like, I can post a Python variant of it.)

I decided not to wait for your answer... :)

def distance(a,b):
    c = {}
    n = len(a); m = len(b)
    for i in range(0,n+1):
        c[i,0] = i
    for j in range(0,m+1):
        c[0,j] = j      
    for i in range(1,n+1):
        for j in range(1,m+1):
            x = c[i-1,j]+1
            y = c[i,j-1]+1
            if a[i-1] == b[j-1]:
                z = c[i-1,j-1]
            else:
                z = c[i-1,j-1]+1
            c[i,j] = min(x,y,z)
    return c[n,m]

This calculates the minimal number of "operations" - where the
operations are adding, deleting, or translating a character - that are
needed to get from the string a to the string b. This, then, can be
used as a distance measure...

For instance

  distance("big card of doom","bigg chard ov duum") == 5

  distance("big card of doom","little card of phlegm") == 10

You could combine this with other operations, like making everything
lowercase, splitting into separate words and trying to look them up
(perhaps fuzzily) in a word-list - perhaps you could even do a
distance measure on the list-of-words in the same way...

I guess the simplest way to use it would be to just test it against
*all* the cards you have stored, and selecting the ones that got the
lowest distance as likely candidates... There might be some more
efficient algorithm, though...

(And if anyone wants an explanation as to how this distance function
really works, I'll be glad to supply it... I've done lectures about it
in an algorithm class - and will be doing so later this fall as well...)

> 
> > 
> > TIA,

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From crdempsey at uswest.net  Thu Aug 26 23:24:36 1999
From: crdempsey at uswest.net (Syniq)
Date: Thu, 26 Aug 1999 21:24:36 -0600
Subject: Newbie Needs Direction
References: <37C3BC8F.1DFFC151@megsinet.net>
Message-ID: <37C604F4.25A629A2@uswest.net>

David:

For free info--www.python.org

For something you can carry around with you, I would suggest the good old
O'Reilly books:

Python Pocket Reference---for old hands at programming (C/C++)
Programming Python--for those who know a thing or two about programming,
but not much
Learning Python--for those who don't know a thing about programming

I fall into the second group, but I own all three books and find each
useful in its own way.  If you do order them, be sure to check for a link
from python.org--not only do you get your books, but it helps out python!

Cody aka Syniq
crdempsey at uswest.net



David Oppenheimer wrote:

> I have decided to learn python and Tcl.  Are there any great tutorials
> out there for python?  I would also appreciate links to any resources
> you may know about that are full of add-on tools, etc for python.  I am
> also interested in wxPython.  I would like to also know what kinds of
> programs people are designing with python.  And your comments on the
> strengths and/or weaknesses you perceive in the language.   What
> compilers are compatible, etc.?
>
> Thanks,
>
> David O.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: crdempsey.vcf
Type: text/x-vcard
Size: 267 bytes
Desc: Card for Syniq
URL: <http://mail.python.org/pipermail/python-list/attachments/19990826/00f75f75/attachment.vcf>

From chinaski_py at my-deja.com  Sat Aug 21 17:27:37 1999
From: chinaski_py at my-deja.com (chinaski_py at my-deja.com)
Date: Sat, 21 Aug 1999 21:27:37 GMT
Subject: Thanks Hever, but what about...?
Message-ID: <7pn5k1$5e7$1@nnrp1.deja.com>

First,Thanks for the help.I have also posted a similar message at the
Python.org Tutor, but no response.Your correction to my path seemed to
work.I use PythonWin with no trouble, but If I go to Explorer to open a
.py file that I made in PythonWin, The Python "DOS" type screen opens.
Should I be seperating these .py and .pyw? Am I missing something by
just using PythonWin?    Also....
What about the "Idle"(Python GUI).When I click on it in the Python file
(Windows box pops up) it says:
PROGRAM NOT FOUND
Windows can not find Program.exe.

This program is needed for opening files of
type 'Python File (no console)'.
Location of Program.exe:
(box)
C:\ -What goes here?
                      Your help is appreciated and good karma always
comes back to you,
                     G.Norton (chinaski_py)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From m.faassen at vet.uu.nl  Tue Aug 31 05:32:04 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 31 Aug 1999 09:32:04 GMT
Subject: Parrot: Why the clunky syntax?
References: <935279776snz@vision25.demon.co.uk> <935454399snz@vision25.demon.co.uk> <37C2B9CE.95AA5FEC@lri.fr> <37C9FF53.DD466783@compaq.com> <19990830150321.C6858@optiplex.palga.uucp>
Message-ID: <7qg7ek$k5e$1@newshost.accu.uu.nl>

Gerrit Holl <gerrit.holl at pobox.com> wrote:
> On Mon, Aug 30, 1999 at 03:49:39PM +1200, Greg Ewing wrote:
[ { } syntax versus python style indenting syntax example]
>> After all, we like Python because it does without
>> line noise like @{} -- don't we?
>> 

> What are you talking about?
> I think I missed something...

Many programming languages (such as C, C++, Perl and Java) use { and } to
mark block structure. So, instead of Python's:

if 1 == 1:
    print "We're glad 1 is still constant!"

you can write (morphed Python code -- this isn't actual C or anything):

if 1 == 1: {
    print "We're glad 1 is still constant!"
}

But here, indentation is only for readability; it doesn't have any special
meaning. So you could write this too:

if 1 == 1: { print "We're glad 1 is still constant!"
}

or whatever.

So what Greg was asking, why not use a more Pythonesque syntax for parrot,
where indentation has actual meaning?

And it's a good question. :)

Regards,

Martijn




From barmar at bbnplanet.com  Thu Aug 19 20:24:25 1999
From: barmar at bbnplanet.com (Barry Margolin)
Date: Fri, 20 Aug 1999 00:24:25 GMT
Subject: 'if' as a _function_, in Scheme and Python
References: <7pi0kl$ko8$1@nnrp1.deja.com> <37bc9dd6.11041486@90.0.0.40>
Message-ID: <Ze1v3.287$m84.4462@burlma1-snr2>

In article <37bc9dd6.11041486 at 90.0.0.40>,
Steve Schafer <pandeng at telepath.com> wrote:
>On Thu, 19 Aug 1999 22:31:49 GMT, oleg at pobox.com wrote:
>
>>To show it clearer, let me re-write my-if as
>>
>>        (define (my-if condition then-branch else-branch)
>>           ((cdr (assq (eq? condition #f)
>>                (list (cons #t else-branch) (cons #f then-branch))))))
>>
>>This definition is _purely_ functional
>
>Not only is it purely functional, it also unconditionally evaluates
>both branches (in Scheme). Therefore, it is not equivalent to IF.

No.  Remember that then-branch and else-branch are thunks.  It doesn't
matter whether you evaluate them, they don't do anything until you *call*
them.  The above expression selects one of them and then calls it (notice
the extra level of parentheses before "cdr").

The subject of a functional "if" comes up in the Scheme group every 6-9
months, I think.  The above implementation is clever.  A more common
implementation involves replacing #t and #f with functions:

(define (if-true then-branch else-branch)
  (then-branch))
(define (if-false then-branch else-branch)
  (else-branch))

Predicates return if-true or if-false instead of #t or #f, respectively (we
could define that these special values actually evaluate to these
functions).  Then "if" is defined as:

(define (if condition then-branch else-branch)
  (condition then-branch else-branch))

-- 
Barry Margolin, barmar at bbnplanet.com
GTE Internetworking, Powered by BBN, Burlington, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.



From Thornem at labcorp.com  Thu Aug  5 09:23:15 1999
From: Thornem at labcorp.com (Michael Thorne)
Date: Thu, 05 Aug 1999 09:23:15 -0400
Subject: Looking for a "Tip of the Day"
Message-ID: <s7a9581d.096@lcagw.labcorp.com>

Hi All,

  TipWorld at http://www.tipworld.com supplies a variety of forums that
will email out a short tip each day on the given subject.  Being a newbie
to LINUX and DHTML I subscribe to these two and it's very nice to get
this every day and slowly learn.  Being brand new to Python, on page 56
of "Learning Python", it would be great to have the same service available
for this cool language.

  Is there an "Anaconda" out there who could take on this effort - it would
be much appreciated!

Thanks in advance,


Michael.

Dallas.TX.US


THINK
Think Different
Think Open Source




From hei at adtranzsig.de  Wed Aug 11 05:53:02 1999
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Wed, 11 Aug 1999 11:53:02 +0200
Subject: Weird problem ?
References: <000101bee1ca$e9795940$46f8143e@pc58-mikael.fas.flygt.com>
Message-ID: <934367506.892011@news.adtranzsig.de>

Hi Mikael!

Besides the problem you asked for, your code has
the problem of looking a bit confusing:
>
>class VECTOR:

The following three lines make no sense,
or at least not the sense you had in mind,
i suspect:
Here, you set up three variables that
could be accessed with "VECTOR.x",
"VECTOR.y", "VECTOR.z" -
they don't belong to any instance.

>   x = 0.0
>   y = 0.0
>   z = 0.0

Now, you let each instance create
its own x,y,z (which is presumably what you
wanted to do):
>   def __init__(self,xx=0.0,yy=0.0,zz=0.0):
>      self.x = xx
>      self.y = yy
>      self.z = zz

You don't have to declare variables in Python.
the 3 lines under "class VECTOR" don't do what
you wanted them to, just delete them.
--
Dirk
Where did you want to go again?





From tchrist at mox.perl.com  Thu Aug 19 14:38:25 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 11:38:25 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <37bc4111@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Michael Hudson <mwh21 at cam.ac.uk> writes:
:> GOTCHA: (high)
:>     Local variables are never declared.
:
:*Nothing* is ever delcared in Python.

This is hardly a happiness. :-(

I've been finding out how much more that Perl's compiler does with symbols
than Python's does.  For example, in Perl, a sub fn(){} is compiled when
the compiler sees it, and goes in the symbol table then.  In Python,
it appears that def fn(): is a run-time event, and executable statement.
Hence the need to get all your defs in first.  Perl's declarations are
sub, format, package, and my.

Another difference is that because the Perl compiler does the symbol table
lookup at compile time for both globals and locals, including 
functions.  Indirect run-time lookups that happen at run-time are
slower.

:Well, no. How would you like to pass a tuple to a function?

I'd like using

    fn(t)

and then the function deciding whether it got enoguh arguments.

: 
:> GOTCHA: (medium)
:>     Perl's hex() function is the opposite of Python's.  Perl's hex
:>     function converts 22 into 37.  Python's hex function converts 37 into
:>     0x22.  Oh my!  Like hex, Python has oct() completely backwards from
:>     Perl.  Python's oct(44) returns 054, but Perl's oct(44) return 36.
:
:This would have to do with strings not being automatically coerced
:into numbers, methinks.

No, that's not it.  Perl's hex takes a hex number and returns 
a decimal one; same with oct.  Python's is the reverse.  Very weird.

    echo 'print oct(12)' | python
    14

    echo 'print oct(12)' | perl -l
    10

See what I mean?

:There's info format if you look for it and you have to have a CLI.

At the very least, python needs a top-level python(1) manpage that gives
calling conventions, options, envariables etc, and an explanation of how
to locate the rest of the documentation (even if it's in non-man form).
That's a POSIX.2 requirement.

And info sucks.  Please don't emacs me.  I hurt enough already.

:
:> GOTCHA: (low)
:>     Often Python's error messages leave something to be desired.
:>     I don't know whether 
:
:Is this a joke? This is improving.

No, it isn't a joke.  The error messages are often mysterious.
Contrast:

% perl -e 'fn()'
Undefined subroutine &main::fn called at -e line 1.

% echo 'fn()' | python
Traceback (innermost last):
  File "<stdin>", line 1, in ?
NameError: fn

:> 	print s[2:4]
:>       ri
:>     Yes, that's all you got.  Strange, eh?  See below on ranges.
:
:What were you expecting?

Charcters 2 through 4, of course.  Hey, does python do Unicode
yet?

:> GOTCHA: (medium)
:>     You can't slice dictionaries at all in Python.  In Perl, it's
:>     easy to slice a hash, and just as sensible.
:
:Huh? What would that mean? There's no reason to suspect that the keys
:of a dictionary are ordered.

Order is irrelevant.  You seem to be equating slicing with ranges.

    @a = ("this", "and", "that", "are", "all", "ready")
    print "@a[-1, 2, 4, 2, 0, -2]\n"

Which prints:

    ready that ready that this are

Likewise, with a hash.  For example:

    print "@ENV{HOME,TERM,USER}\n";

Which prints:

    /home/tchrist xterm tchrist

So slicing is just selecting multiple values at once.  I can assign
to a hash that way, too:

    @wife_of{"fred", "barney"} = ("wilma", "betty");

:new = old.copy()

I wonder how many times I'll have to explain that I read the 
Lutz book, wasn't on the Internet at the time, and found no manpages
installed with python on my laptop.

:>     There's no way to set up a permitted exports list.  The caller may have
:>     anything they ask for.  
:
:This is a general Python convention; you *can* shoot yourself in the
:foot if you try hard enough. Means the debugger can be written in
:Python, for one thing.

We did that in Perl, too, nevertheless. 

:Yes this is a bit annoying. The Right Way to attach state to a
:function is a class, of course. Not that that helps much.

That sounds like religion.  And aren't bound methods inherent
memory leaks?

    fn = obj.method;

:> GOTCHA: (low)
:>     Python's eval() only works on expressions, not full code blocks full
:>     of statements.  You need exec() for the whole thing.
:
:Writing exec() suggests it's a function; it isn't, it's a statement.

Funny that you guys are all flustered about this statement/function thing.
I think of if and def as statements.  I'm not saying print and exec
aren't, but they sure look just like it.
:
:> COOLNESS:
:>     I believe that Python checks the prototype signature on methods
:>     as well as on functions.  I wonder whether Tim Bunce's ideas run-time
:>     evaluation of prototypes for Perl might be able to do this.
:
:You mean perl doesn't? Ouch.

Method calls go through the symbol table at run-time.  Function calls
are resolve at compile time, and need no symbol table stuff.  That means
you get prototype errors at compile time, before you start, if you get
them at all.  But because we don't know what type is going to be in
the object, the compiler can't know, and the interpreter doesn't care.
Well, it cares and will hurl an exception if it can't find the method
by that name, but it does no proto checks.  This may change.

Technically, there are the new optional lexical declarations to help
the compiler and the backend C code-generator, but those still leave a
lot to be desired.

--tom
-- 
To stay young requires unceasing cultivation of the ability to unlearn
old falsehoods.                                        -- Lazarus Long



From ScherBi at BAM.com  Tue Aug 17 13:17:30 1999
From: ScherBi at BAM.com (ScherBi at BAM.com)
Date: Tue, 17 Aug 1999 13:17:30 -0400
Subject: UML
Message-ID: <3F8BA119B48AD11197470008C7A493CB98B502@HQOBGEX1>

I use UML, via ObjectDomain (www.objectdomain.com).
I am involved in developing web based applications for our intranet.
I find it extremely useful for most anything larger than a quick hack.

Where I find it most useful:

 - Use Case diagrams are great for capturing requirments.
 - UML allows one to work on 'the big picture', without worrying much about
the details, then later you can work on individual details without worrying
about the big picture. In other words, create classes and class diagrams in
UML, design the class interfaces, and play around with the design at this
level.  When you think you have it right, generate the stub code and fill in
the methods.  You can fill in the methods without getting caught up in the
interface issues, which makes things much easier.

You can download a 30 day demo from objectdomain.  To help learn UML, I
recommend getting a python module (that has a class hierarchy in it, like
SocketServer.py) where you understand the hierarchy and the operations, then
reverse engineer it into objectdomain.  You can then automatically generate
a class diagram.  

And no, I don't work for ObjectDomain.  It's the only UML tool I know of the
generates and reverse engineers Python.  OD is written in Java.  I don't
like that feature of it much, but once I get over the problems inherint in
that, I think it's a fine tool.

It's rumored that third parties develop language modules for Rational Rose,
others may know more about this.


Bill Scherer

-----Original Message-----
From: robert_wilhelm_land [mailto:robert_wilhelm_land at compuserve.com]
Sent: Tuesday, August 17, 1999 9:57 AM
To: python-list at python.org
Subject: UML


Are any persons in this group using the unified modelling language?
If yes, could you kindly explain when to use the UML and what the
advantages are?
I have read some articles, but I'm not able to analyse the sense of the
UML.

Would be grateful for some help


Robert



-- 
http://www.python.org/mailman/listinfo/python-list



From fdrake at cnri.reston.va.us  Thu Aug  5 11:07:39 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake, Jr.)
Date: Thu, 5 Aug 1999 15:07:39 GMT
Subject: [Image-SIG] Proper application of the buffer interface
In-Reply-To: <006501bedeee$33ef5c20$1101a8c0@bobcat>
References: <14248.63286.73014.204886@weyr.cnri.reston.va.us>
	<006501bedeee$33ef5c20$1101a8c0@bobcat>
Message-ID: <14249.43195.9095.338176@weyr.cnri.reston.va.us>

Mark Hammond writes:
 > The good news is that it is most common for extension modules to spell
 > "give me a string with the raw binary data" as "PyArg_ParseTyple("s#",
 > ...);".  PyArg_ParseTuple has been upgraded to use the buffer interfaces,

Mark,
  This is really cool.  Buffers are probably what I want, then; the
question is starting to become whether to use a buffer object or to
create a new object that implements the buffer interface.  I think I
can handle that!  (And it's looking like implementing a new C type,
for those following t1python.)

 > The problem remains, of course, for extensions that use the
 > PyString_Check(), PyString_AsString() etc functions.  If they were upgraded
 > to use the buffer interfaces, then the transition would be complete.

  Understandable; this is the price of using the lowest-level concrete 
object interfaces.

 > strings for this purpose, but clearly indicate the data is not really a
 > string!  This appears to be more a matter of style, and also paves the road
 > to Unicode - eg, it makes sense to convert any Python string object to
 > Unicode, but not necessarily a binary buffer.

  This is a compelling argument for the buffer type/interface in my
book.  I won't be happy until Unicode strings are in the core!  (No,
I'm not holding out for Unicode strings to replace the current string
type, just that they be in the core and well-supported.)

 > Probably Greg and Guido are the only 2 with the real insight, as they
 > threashed out the details.  But Im pretty happy with my understanding (as
 > detailed above) of the issues.

  Your elucidation on the topic is excellent; if you'd like to write a 
section for the "Extending & Embedding" manual regarding when to
implement the buffer interface and when to use it, I'd certainly be
glad to mark it up and integrate it.  ;-)
  (Greg: No, this wouldn't get you off the hook for reference material!)


  -Fred

--
Fred L. Drake, Jr. <fdrake at cnri.reston.va.us>
Corporation for National Research Initiatives



From HellPhyre at iname.com  Mon Aug  2 12:54:41 1999
From: HellPhyre at iname.com (HellPhyre at iname.com)
Date: Mon, 02 Aug 1999 12:54:41 -0400
Subject: [Tutor] I want to learn to hack
References: <19990721154610.B3663@optiplex.palga.uucp> <Pine.WNT.4.04.9907211133150.294-100000@rigoletto.ski.org> <14231.11726.916541.492153@nobot.cnri.reston.va.us> <slrn7q1dau.17t.esr@snark.thyrsus.com>
Message-ID: <37A5CD51.3880D558@iname.com>

> Some time back, I published a page called "How To Become A Hacker"
> <URL:http://earthspace.net/~esr/faqs/hacker-howto.html">.  As a result,
> I get a couple dozen semi-illiterate email notes a week from teenagers
> wanting me to teach them everything I know.  Instantly.

Well, I have been a hacker for a good 7-8 years now, and I never quite found
any text that quite caught the essance of hacking quite like that.  After
reading it, I have started reading up on Python.  I can already program in
C, Assmebly, Basic, and (gasp) Pascal.  Anyway I'm just trailing off.  I
wrote a text file a while back, and made the mistake of putting an e-mail
address on it.  From that point on I got 50+ e-mails a day.  It got bad
enough that I finally just deleted the e-mail account and started a new
one.  It seems that some newbies just don't know how to do anything on their
own; which defeats the entire purpose of hacking.




From wlfraed at ix.netcom.com  Sat Aug 28 00:03:18 1999
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Sat, 28 Aug 1999 04:03:18 GMT
Subject: Slow python on particular computer
References: <1276313212-39052155@hypernet.com>
Message-ID: <37c75d62.2975124@nntp.best.ix.netcom.com>

On Fri, 27 Aug 1999 23:05:50 -0500, "Gordon McMillan"
<gmcm at hypernet.com> declaimed the following in comp.lang.python:

> Paul Prescod wrote:
> 
> > Has anyone seen a case where Python runs REALLY, painfully slowly on
> > an otherwise spritely Windows 95 machine? It takes something like 8
> > seconds to launch the interpreted and parse a 10 line program.
> > Something must be mis-configured.
> 
> Disk fragmentation is the most likely problem. Maybe registry 
> pollution, especially if you're using COM.
>
	I'd add a configuration in which ALL "free" RAM was sucked up as
disk cache, which means loading anything new requires first purging
loads of RAM out to disk to make room. (First thing I did at work after
my snail [90MHzPentium] was converted to W95 was run Sysedit and add the
vcache section -- after that it no longer to two minutes to load an
application).

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <



From aarhodes at cisco.com  Mon Aug 23 20:02:19 1999
From: aarhodes at cisco.com (Aaron Rhodes)
Date: Tue, 24 Aug 1999 00:02:19 +0000
Subject: Writing data to a pty file descriptor
Message-ID: <37C1E10B.D2170157@cisco.com>

Pythoners,

I'm trying to write data to a pty (pseudo terminal).
I'm using the pty module (pty.writen) after forking a
child with a controlling terminal, execing the shell, etc.
(similar to what Expect does or the code in Ch.19 of
Stevens' Advanced Programming in the UNIX environment)

I'm not sure why but I can't write certain characters
to the pty. I've written the following command to the pty to set
up for a file input from the shell command line.

$ cat > file.uue << EOF
>

However, as soon as I try to write certain characters,
the shell seems to interpret them incorrectly.  One
character that does this is the back-tick -- `
character.  This seems odd to me since I can input
the characters from stdin just fine... but not by
writing to the pty directly.
Any ideas about why a write to the pty from python
would fail?

Aaron

--
Aaron Rhodes
aarhodes at cisco.com
(512) 378-1097






From famewolf at my-deja.com  Sat Aug 21 14:48:23 1999
From: famewolf at my-deja.com (famewolf at my-deja.com)
Date: Sat, 21 Aug 1999 18:48:23 GMT
Subject: downloading web based email
References: <7och4d$jr0$1@nnrp1.deja.com> <14249.55454.471036.787883@dolphin.mojam.com> <37AA6BCD.1A611445@alum.mit.edu>
Message-ID: <7pms9l$v4o$1@nnrp1.deja.com>

What I really would like to see is a reliable program to let me send
pop3 mail and have it send it through my web based email account...I've
tried cwebmail, webpop15 and zincheck which all say they work with yahoo
mail or with mailcity and none of them will send a mail through
either...I spose their out of date.  Altavista is blocking pop3 outgoing
to "prevent abuse" and it's a major hassle to log into the web email
account everytime I want to send a quick inquiry on someone's web page.


In article <37AA6BCD.1A611445 at alum.mit.edu>,
  pehr at alum.mit.edu wrote:
> Yahoo *definitely* supports reading from their webmail accounts
> though the universal standard of pop3.  Fetchmail will get this,
> or you could simply point your netscape mail client at the yahoo
server.
> To enable this feature under yahoo, you have to click an option which
> allows them to send you adds. Adds pay for their service so if you
don't
> go to their web page they have to be able to still get your attention.
>
> I'm really impressed with how yahoo treats their clients.
> Yahoo mail lets you specify the return address so you aren't
> "locked in" to a yahoo account name, you can have mail forwarded from
> a more permanent mail address and set your reply address to the
permanent one!
>
>     ---pehr
>
> Skip Montanaro wrote:
>
> >     Preston> I'm thinking of writing a Python program to download
email from
> >     Preston> web-based services such as Yahoo mail, Deja, Hotmail,
and so
> >     Preston> on, and deliever the mail in a format that standard
unix
> >     Preston> mailers can use.
> >
> >     Preston> I'm posting here to make sure that this work has not
already
> >     Preston> been done.  If someone has already seen a program like
this,
> >     Preston> that can be adapted for a variety of web based email
services,
> >     Preston> please let me know.  I'm not writing this as a learning
> >     Preston> exercise, I just need the tool. ;-)
> >
> > Presuming the services you're interested in support a standard mail
transfer
> > protocol like POP or IMAP, you need look no further than Eric
Raymond's
> > excellent fetchmail program:
> >
> >     http://www.tuxedo.org/~esr/fetchmail/
> >
> > If not, perhaps you can train fetchmail to use whatever transport
they do
> > offer.
> >
> > Skip Montanaro  | http://www.mojam.com/
> > skip at mojam.com  | http://www.musi-cal.com/~skip/
> > 847-971-7098
>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From philwise at geocities.com  Tue Aug 31 13:59:59 1999
From: philwise at geocities.com (Phil Wise)
Date: Tue, 31 Aug 1999 17:59:59 GMT
Subject: Doing POST opperations from Python
Message-ID: <37cbeeee.7358361@news.freeserve.net>

	I am trying to produce POST requests within python
(to drive a web based email client). I can do GET requests,
but can understand how to do post. I have read RFC1945
(Informational), but I can find the info there either.
I understand that the data is sent as part of the headers,
but how?

----------------------------------------
Spam this address and wave goodbye to 
     your internet connection :
       philwise at geocities.com
----------------------------------------



From news at dorb.com  Tue Aug  3 20:48:16 1999
From: news at dorb.com (Darrell)
Date: Tue, 3 Aug 1999 20:48:16 -0400
Subject: Printing from python on Windows platform
References: <Pine.LNX.3.96.990803232344.254A-100000@Johann>
Message-ID: <H5Mp3.1603$ib.56555@typhoon.nyroc.rr.com>

Printing from Windows isn't so straight forward.
I've seen printing support in wxPython.

In the old DOS days you could write to lpt1:

E:\dgXXX>copy win.cfg lpt1:
        1 file(s) copied.

Nothing came out of my printer.

E:\dgXXX>dir lpt*
 Volume in drive E has no label.
 Volume Serial Number is 541E-3EC8

 Directory of E:\ace\v2\data\dgXXX

File Not Found


E:\dgXXX>copy win.cfg stdout:
"stdout:" is not a recognized device.
The system cannot find the file specified.

I'm using NT 4.0

Hummm.
--
--Darrell
Johann Spies <jhspies at futurenet.co.za> wrote in message
news:Pine.LNX.3.96.990803232344.254A-100000 at Johann...
> I am sorry to ask this a second time, but I did not receive any hint
> after the first time I put this question to the list.
>
> I am a Linux user writing a program for somebody that will use it on a
> Windows platform.  In Linux I use the following to redirect printing to
> the printer:
>
>     drukker = sys.stdout
>     drukker = os.popen('lpr','w')
>     drukker.write(drukstuk)
>     sys.stdout.write(drukstuk)
>
> What would the windows equivalent be?
>
> I have tried os.popen('prn', 'w'), but without success.  I cannot see
> anything in the python documentation to help me.
>
>
> Johann
>
>  -------------------------------------------------------------------------
-
> | Johann Spies                                 Windsorlaan 19
|
> | jhspies at futurenet.co.za                3201 Pietermaritzburg
|
> | Tel/Faks Nr. +27 331-46-1310        Suid-Afrika (South Africa)  |
>  -------------------------------------------------------------------------
-
>
>      "Have not I commanded thee? Be strong and of a good
>       courage; be not afraid, neither be thou dismayed; for
>       the LORD thy God is with thee whithersoever thou
>       goest."                        Joshua 1:9
>
>
>





From cbbrowne at news.hex.net  Thu Aug 12 23:25:33 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Fri, 13 Aug 1999 03:25:33 GMT
Subject: Anyone else find Tk akward?
References: <37B34DBB.E5A79760@sprint.ca>
Message-ID: <NeMs3.6611$gO1.175350@news2.giganews.com>

On Thu, 12 Aug 1999 15:42:03 -0700, Orlando Vazquez
<ovazquez at sprint.ca> wrote: 
>Is it just me, or is the way Tk organizes things on the window someone
>weird? I'm just starting to play around with it, and it has me puzzled.
>Wouldnt it be a lot easier to just use a coordinte system?
>For example put a button at (0,0) with length 10 and width 20 etc..
>Are there any reasons why it _has_ to be this way?

*What* coordinate system?

What units are you using?

The Tk model of packing "objects" onto a "canvas" allows it to
self-optimize, and lets the programmer not need to worry so much about
those sorts of fiddly details.
-- 
"The wrath of Holloway is nothing compared to the wrath of Moon."
-- Fred Drenckhahn
cbbrowne at hex.net- <http://www.hex.net/~cbbrowne/lsf.html>



From tchrist at mox.perl.com  Thu Aug 19 16:05:24 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 13:05:24 -0700
Subject: More random python observations from a perl programmer
References: <199908191713.LAA02152@jhereg.perl.com> <5l4shvtxfe.fsf@eric.cnri.reston.va.us>
Message-ID: <37bc5574@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Guido van Rossum <guido at cnri.reston.va.us> writes:
:Please let the issue go.  As we say, you are welcome to fix it
:yourself.  We'd even accept a Perl script that turns our latex (soon
:to be SGML or XML) into man page format.

I tend to answer every criticism instead of letting it settle
in and saying one answer once.  This is sub-optimal.

Your wishes shall be granted--if I can; I'll certainly try.

:Since you're an O'Reilly author, I'm surprised you didn't base your
:(otherwise excellent) Perl/Python guide on the much more recent (April
:99) Learning Python by Mark Lutz and David Ascher.  It would have
:saved you a whole bunch of pitfalls.

I couldn't find it when I was looking on the way to the desert.  I could
only find the first one, so that's the one I took with me!  I'm not sure
whether I have the other one still, or whether one of my random visitors
absconded with it.

--tom
-- 
"All mainstream science was once heresy."
				- Gould



From tchrist at mox.perl.com  Fri Aug 20 05:02:31 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 20 Aug 1999 02:02:31 -0700
Subject: making python man pages looks hard
References: <1103_935134343@shoal>
Message-ID: <37bd0b97@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    connolly at w3.org (Dan Connolly) writes:
:parting shot: CPAN is cool, but I find it frustrating that I can't
:read the documentation for a module without downloading
:and upacking the module. for example, I can browse
:the list of modules,
:
:http://www.perl.com/CPAN/modules/00modlist.long.html

Try

    http://search.cpan.org/

--tom
-- 
You have to admit that it's difficult to misplace the Perl sources.  :-)
		--Larry Wall in <1992Aug26.184221.29627 at netlabs.com>



From jstevens at bamboo.verinet.com  Fri Aug 13 23:08:33 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: Sat, 14 Aug 1999 03:08:33 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ce93@cs.colorado.edu> <slrn7r9la9.pu8.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <slrn7r9nde.h3c.jstevens@bamboo.verinet.com>

On 14 Aug 1999 02:32:12 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>On 13 Aug 1999 20:04:03 -0700, John W. Stevens <jstevens at basho.fc.hp.com> wrote:
>>> In comp.lang.perl.misc, 
>>>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
>>> :$b[ 2 ] = $c;
>>> :
>>> :> That's just fine in Perl.  It's not fine in Python, because Python
>>> :> won't automatically grow an array.
>>> :
>>> :'Cause it doesn't have arrays (or, at least, not built in ones).
>>> 
>>> Gosh, that's a feature.  NOT.
>>
>>Perl doesn't have lists.  Python doesn't have built-in arrays.
>
>You should learn some perl you now..
>
>@array = (1,10,20,30);
>$from_list = (1,10,20,30);
>$from_array = @array;
>print "$from_list\n$from_array\n";
>
>Will output :
>30
>4

The @ prefix denotes an array.  You, yourself, should learn
Perl.  Calling an array a list, doesn't make it one.

>Perl has lists,

Not built in, it doesn't, unless you define array and list as being
different words for exactly the same type/class.

>if you know perl you would know this.

I know Perl.  You need to learn Python.

>If you program in perl 
>and don't know this, then you must get very very confused at times.

If @ denotes list, then the following Perl would be illegal:

@ary = (1, 2, 3);
@ary[5] = "Test";

But, obviously, this is not illegal.

>>I will assume that a list module is available for Perl.
>
>No it is one of the built in bits... like hashes and arrays.

Really?  What is the prefix character that denotes a list?

>>I wasn't trying to compare features, I was simply pointing out
>>that your comparison was Apples and Oranges, and therefore at
>>least somewhat invalid.
>
>Only because you have no idea what you are talking about.

:-)

Coming from somebody who doesn't know the difference from *EMULATING*
a list with an array, vs. a real array, that is a good one!

Now I suppose that you will tell me that Perl has stacks, too!
;->

John S.



From fl_aggie at thepentagon.com  Sat Aug 14 22:55:38 1999
From: fl_aggie at thepentagon.com (I R A Darth Aggie)
Date: 15 Aug 1999 02:55:38 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
Message-ID: <slrn7rcb5c.5k9.fl_aggie@thepentagon.com>

On Sat, 14 Aug 1999 17:39:51 GMT, thehaas at my-deja.com <thehaas at my-deja.com>, in
<7p49l5$uv7$1 at nnrp1.deja.com> wrote:

+ I've been reading that flame war

I've skipped the latest installment of 142 posts...

+ but I noticed that it is the Perl
+ people, for the most part, that have donned the asbestos undies and
+ charged up their flame throwers

Ummm...with a subject line like "Why use Perl when we've got Python?"
Or would you say that such a subject is non-inflamatory?

James

-- 
Consulting Minister for Consultants, DNRC
The Bill of Rights is paid in Responsibilities - Jean McGuire
To cure your perl CGI problems, please look at:
<url:http://www.perl.com/CPAN/doc/FAQs/cgi/idiots-guide.html>



From I.Clarke at strs.co.uk  Mon Aug 16 06:02:20 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Mon, 16 Aug 1999 11:02:20 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com>
Message-ID: <37B7E1AC.FF4FF2CF@strs.co.uk>

> The problem in comp.lang.perl.misc is not the beginners -- it is
> the regulars.

I agree completely.  As the person who started this whole thing
(sheepish grin) the regulars, Tom in particular, has been rude, and
resorted to personal insults on a number of occasions, even while I
remained calm and polite.  He repeatedly accuses me of being a troll and
inciting flames (despite the fact that he was the only one flaming!). 
To top it all off, having posted numerous "courtesy" copies of his
newsgroup postings to my email address, to which I replied on-subject
and politely, he then emailed me an email saying that I had been
killfiled!  A rather pathetic way to get the last word in my opinion.

Lucky I do not judge Perl on the quality of their newsgroup regulars.

Ian.



From cjc26 at nospam.cornell.edu  Wed Aug 18 19:16:42 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 18 Aug 1999 23:16:42 GMT
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <y0jiu6edoa9.fsf@vier.idi.ntnu.no> <xj0d7wl9acd.fsf@cyclone.harvard.edu>
Message-ID: <slrn7rm1ol.s1.cjc26@tankgrrl.bridget.mindriot.net>

Pada 17 Aug 1999 20:54:58 -0400, Michael Haggerty bilang:
| mlh at idt.ntnu.no (Magnus L. Hetland) writes:
| 
| > result = []
| > for i in range(len(l)):
| >     element = random.choice(list)
| >     l.remove(element)
| >     result.append(element)
| > 
| > Probably not that quick, but...
| 
| Indeed not that quick.  This shuffle is O(N**2) because each time
| through the loop it has to (1) search for the element by value and (2)
| delete the element which involved shifting the rest of the list over
| by one.

I don't know how lists are implemented in Python, but if they're linked lists
instead of arrays, then deleting an element (once you've already found it)
should take constant time.  But then the append will be O(n).


| Each of those actions is O(N) so the whole loop is O(N**2).

Nope, it should still be O(n).


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From mwh21 at cam.ac.uk  Sat Aug 28 10:31:33 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 28 Aug 1999 15:31:33 +0100
Subject: A Plea For A Return To Sanity!!!
References: <000301bef12c$1bfd6940$ad2d153f@tim> <0fbd01bef154$4f7ab020$0301a8c0@mnot.net>
Message-ID: <m37lmg0yey.fsf@atrus.jesus.cam.ac.uk>

"Mark Nottingham" <mnot at mnot.net> writes:

> > python-folk-are-the-best-buddies-psu-dues-can-buy-ly y'rs  - tim
> 
> I don't know why, but I'm suddenly consumed with the desire to pay my PSA
> dues... Hmmm...
> 
> Seriously, I've had a printed-out PSA application sitting under a pile of
> paper on my desk for the better part of a year. Why? Well, have you ever
> tried to post a check (or as they say here, cheque) from Australia to
> Reston, VA? It's a pain, unreliable, and financially challenging.

You can pay by credit card - that was fairly straightforward from here
(UK).

> Has anyone over there looked into one of the online payment services like
> Kagi, etc? It would make things a lot easier, and you might get a few more
> members. Like me. They take a few percent, but it's pretty reasonable.

That would be cool too.

Regards,
Michael



From jonathanbrisbin at excite.com  Thu Aug 26 16:46:07 1999
From: jonathanbrisbin at excite.com (jonathanbrisbin at excite.com)
Date: Thu, 26 Aug 1999 13:46:07 PDT
Subject: Python article on Builder.com
Message-ID: <935700367.20024.603@excite.com>

Spam detection software, running on the system "albatross.python.org", has
identified this incoming email as possible spam.  The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email.  If you have any questions, see
the administrator of that system for details.

Content preview:  On Thu, 26 Aug 1999 14:34:11 -0400, Tom Funk wrote: > Some
   unanswered issues have been raised that could use some input from > some
  experienced Python developers. Also, some comments from someone > who's used
   both Perl and Python would be welcome (only one such reader > has chimed
  in to date). > [...] 

Content analysis details:   (5.3 points, 5.0 required)

 pts rule name              description
---- ---------------------- --------------------------------------------------
 2.0 FH_DATE_IS_19XX        The date is not 19xx.
 0.8 NO_RDNS_DOTCOM_HELO    Host HELO'd as a big ISP, but had no rDNS
 2.6 FAKE_HELO_EXCITE       Host HELO did not match rDNS: excite.com


-------------- next part --------------
An embedded message was scrubbed...
From: jonathanbrisbin at excite.com
Subject: Re: Python article on Builder.com
Date: Thu, 26 Aug 1999 13:46:07 PDT
Size: 2109
URL: <http://mail.python.org/pipermail/python-list/attachments/19990826/c042cf06/attachment.mht>

From philh at vision25.demon.co.uk  Tue Aug 17 14:26:54 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 17 Aug 99 18:26:54 GMT
Subject: CVS classes in Python
References: <37B9969B.BB40C25F@primenet.com>
Message-ID: <934914414snz@vision25.demon.co.uk>

In article <37B9969B.BB40C25F at primenet.com>
           yerny at primenet.com "Bryan Jernigan" writes:
> I'm a newbie to Python and I would like to create a Python script to
> maintain a distributed CVS CM system.  Before I re-invent the wheel, I
> was wondering if anyone has some decent CVS classes and/or functions.
> If so, please drop me a line.

Why not have your Python classes acting as a front end to CVS or PVS?

-- 
Phil Hunt....philh at vision25.demon.co.uk




From robin at jessikat.demon.co.uk  Fri Aug  6 10:49:03 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Fri, 6 Aug 1999 14:49:03 GMT
Subject: Linux why not fully dynamic?
In-Reply-To: <14250.62239.580394.67647@buffalo.fnal.gov>
References: <aOAnwBA2rjq3EwnZ@jessikat.demon.co.uk>
 <14250.62239.580394.67647@buffalo.fnal.gov>
Message-ID: <HrAQdAAfXvq3Ewmx@jessikat.demon.co.uk>

In message <14250.62239.580394.67647 at buffalo.fnal.gov>, Charles G
Waldman <cgw at fnal.gov> writes
>Robin Becker writes:
> > Why doesn't ./configure set up for a libpython.so.1.5.2 or whatever?
>
>This is an FAQ, see:
> 
> http://www.python.org/doc/FAQ.html#3.30
>
Thanks. 

-- 
Robin Becker



From samschul at pacbell.net  Tue Aug 31 23:44:59 1999
From: samschul at pacbell.net (Sam Schulenburg)
Date: Wed, 01 Sep 1999 03:44:59 GMT
Subject: How to sell Python to your boss?
References: <11A17AA2B9EAD111BCEA00A0C9B41793034AAD03@molach.origin.ea.com>
Message-ID: <7qi7fo$pfs$1@nnrp1.deja.com>

I would try to work out a compromise with your boss. Present him/her
with the idea that you will prototype the project with Python as a
proof of concept, and the rewrite the code in C/C++. I have found that
this approch produces faster more effecient C/C++ code.

Sam Schulenburg


> -----Original Message-----
> From: Petri Mikael Kuittinen [mailto:eye at niksula.hut.fi]
> Sent: Thursday, August 19, 1999 11:54 AM
> To: python-list at cwi.nl
> Subject: How to sell Python to your boss?
>
> I proposed an interesting idea to my employer for my final thesis.
>
> My proposed final thesis includes:
> - a prototype implementation of networked multi-player video gambling
> game (which also has a single-player mode)
> - a design of a network protocol to allow this and automated testing
> of the game (including several automated test player programs which
> have different kind of AI). This protocol and test system is designed
> so that it can easily be expanded to handle any sort of gambling game
>
> Most of the work is design, documentation and programming a
> rudimentary, but easy-to-use, GUI for the multi-player game. My boss
> otherwise agreed with my idea ("it would be really useful..."), but he
> doesn't want me to do the prototype implementation using Python. He
> said I must use a program language that all the other programmers in
> our R&D department understand. Well, that leaves only one choice: C
>
> Some of my game programmer colleagues know a little bit of C++, Java
> and Perl, but these languages cannot be used either. Operating system
> or CPU power is not a problem here (I can use Linux and several
> hundred MHz of CPU power).
>
> I have lots of programming experience in C, and much less in
> Python. Despite this I am confident that I would be able to implement
> the prototype using Python many times faster than using C. Sure I can
> do it with C, or even with assembler (I have made a GUI with m68k
> asm), but parsing text strings and prototype GUI design is so
> laborious in C, that I would only be able to implement a tiny portion
> of the whole thing.
>
> I can understand my bosses opinion. He wants to protect the interest
> of this firm. If I would leave the firm, then someone else would have
> tp spend time learning Python in order to modify my program code.
>
> For several months I have tried to get my colleagues excited to learn
> Python, but so far nobody has been interested to learn new languages,
> except for some mild interest towards Java (which is IMHO an
> overly-hyped language).
>
> Petri
>
> --
> <(O)> Petri Kuittinen, also known as Eye, Dj Eye or
Peku               <(O)>
> <(O)> ADDRESS: Timpurinkuja 1 C 39, FIN-02600 Espoo,
Finland           <(O)>
> <(O)> EMAIL: eye at iki.fi WWW: http://www.iki.fi/~eye/ PHONE: 09-
5472380 <(O)>
> ~The only way to avoid mistakes is to do nothing or try not anything
new.
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fredrik at pythonware.com  Sun Aug 15 06:37:28 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 15 Aug 1999 10:37:28 GMT
Subject: Why use comp.lang.python when you've got comp.lang.perl.misc?
References: <37B5CB95.3FCF571F@appliedbiometrics.com> <199908142005.OAA23215@jhereg.perl.com> <m27lmyj8nv.fsf@netmeg.net>
Message-ID: <009301bee70a$2cf3ffb0$f29b12c2@secret.pythonware.com>

> After reading through the _Programming Python_ book, I thought to
> summarize some of my observatations, but the current climate is just
> too hot. 

maybe Tom lives in Beverly Hills these days? ;-) ;-) ;-)

http://www.amazon.com/exec/obidos/ASIN/1565924649

</F>




From gmcm at hypernet.com  Sun Aug 22 19:46:17 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 22 Aug 1999 18:46:17 -0500
Subject: can some one help a newbie?
In-Reply-To: <slrn7s0bb5.e0r.kaz@ashi.FootPrints.net>
Message-ID: <1276760787-12130790@hypernet.com>

Kaz Kylheku wrote:
> On Sun, 22 Aug 1999 16:44:20 GMT, Fredrik Lundh
> <fredrik at pythonware.com> wrote:
> >a friend working for a big consultancy firm claims
> >they're handing out the following questionnaire to
> >anyone who happens to walk by outside their
> >office:
> >
> >1. have you ever seen a computer: yes/no
> >
> >2. if you answered yes, have you seen a
> >computer in real life (and not just on tv,
> >in a magazine, etc): yes/no
> >
> >one yes and you get a job, two yes and you're
> >hired as an expert consultant.
> 
> What if the one and only yes is in response to question 2? 
> 
> Let me guess, you go straight into management.

No, sales & marketing. 

Management requires 2 nos.

- Gordon



From skaller at maxtal.com.au  Fri Aug 20 19:53:12 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Fri, 20 Aug 1999 23:53:12 GMT
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu> <37BCA74C.D3DCF5DD@fnmail.com>
Message-ID: <37bde8c9.804283468@news.triode.net.au>

On Fri, 20 Aug 1999 02:54:36 +0200, Per Kistler <kistler at fnmail.com> wrote:

>Tom Christiansen wrote:
>> The fact that they did icky windows help and not man pages says
>> something significant.
>
>Why not create something like perldoc for python?
>There must be others missing it...

	Such tools exist, but Pythons docstrings
are not as well principled as Perl's perldod. 

	OTOH, the most powerful literate programming
tool available, interscript, can be used to document python (or perl)
and it makes perlpod (and javadoc) look a bit micky mouse: it is written
in python, can be extended 'indocument' in python,
and it supports perlpod as well :-)

	So in some sense Python offers two options,
'unstructured doc strings' -- which are available at run time --
and fully blown book/web site generation. Perlpod and
javadoc sit in between (neither is 'online', and neither
is powerful enough to generate 'real' documentation)

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From skip at mojam.com  Fri Aug  6 12:22:07 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri,  6 Aug 1999 11:22:07 -0500 (CDT)
Subject: load .py file in interpreter?
In-Reply-To: <37AA3BEE.9417D35A@go2net.com>
References: <37AA3BEE.9417D35A@go2net.com>
Message-ID: <14251.2930.815069.930165@dolphin.mojam.com>

    Joe> Once you're running in the interpreter, how do you load (or
    Joe> re-load) some foo.py Python code into the interpreter? (I'm using
    Joe> JPython, by the way.)

Joe,

Check out the import statement and the reload function as well as the
execfile function you found.  Both import and reload are used to import (or
reimport) a module into the current namespace.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098




From bellings at badlands.NoDak.edu  Thu Aug 19 13:32:36 1999
From: bellings at badlands.NoDak.edu (Brent A Ellingson)
Date: 19 Aug 1999 17:32:36 GMT
Subject: PIGgies?
References: <19990818171523.A22207@digex.net> <37BB6216.55DC0E79@rubic.com>
Message-ID: <7phf3k$72q$1@news.ndsu.nodak.edu>

Jeff Bauer (jbauer at rubic.com) wrote:

: Computer languages used to have meaningful, formidable 
: names in UPPERCASE:  FORTRAN, ALGOL, PL/I.  Now we
: have Python -- named after practitioners of the most
: irreverent form of satire since Lenny Bruce.

See http://www.theonion.com/

--
Brent Ellingson (bellings at badlands.NoDak.edu)
"It is amazing how complete is the delusion that beauty is goodness." 
                                                 -- Leo Tolstoy



From kzaragoza at mediaone.net  Fri Aug 27 23:09:24 1999
From: kzaragoza at mediaone.net (Kris J. Zaragoza)
Date: Sat, 28 Aug 1999 03:09:24 GMT
Subject: Wood, metal and plastic [was: Re: Why do we call python scripting?]
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com> <7q7ccc$5sh44@fido.engr.sgi.com>
Message-ID: <slrn7sekj9.ko.kzaragoza@kzaragoza.ne.mediaone.net>

On 28 Aug 1999 01:01:00 GMT, Paul Jackson <pj at sgi.com> wrote:
>
>    Python is plastic.
>

Great line. Just be careful this doesn't get quoted out of
context. ;-)

>
>(Actually, the way I've heard some of the Lisp crowd speak of
>their various dialects and common usage patterns, that sounds
>pretty plastic to me.  Python differs in having a syntax
>more like the main stream languages, and in not presuming an
>elaborate support environment, both of which make Python much
>more accessible to a main stream programmer, for main stream
>systems, like myself and the systems to which I program.)

It's interesting that you mention Lisp.  I had a conversation recently 
with a friend whom I had introduced to Python.  He had finally taken a 
look and found it quite impressive (particularly JPython).  He did
make one comment which opened my eyes, though: "Now I know why you
like Python so much.  It's Lisp without the parentheses!"

In yet another instance, a friend at work solicited opinions about
Python on a mailing list frequented by ex-Symbolics hackers.  He has
since forwarded me some of the threads that the question touched off.
Many of the comments were positive.  There was, of course, the
required complaint about the use of whitespace as a block delimiter,
but that was to be expected.  My favorite comment was one stating how
absolutely wonderful Python was to embed in other applications.  This
is something that most, if not all Common Lisp systems have lacked.

The more I look at Python, the more I realize how similar it is to
Lisp, if not in syntax, in spirit.  This applies to the whole family
of Lisp languages, not just Common Lisp.  Both Python and Lisp provide
you with a dynamic, interactive environment that doesn't impose on you
any particular programming paradigm.  They strive to provide good
mechanisms for building abstractions to solve tough problems, then
give you maximum flexibility in using and RE-using those abstractions.

Thank you, Guido, for giving us such a great tool to get work done.

-- 
Kris J. Zaragoza           | "Unfortunately, most people can't out-think a
kzaragoza at mediaone.net     | grapefruit."  --Jon Bodner



From mlh at idt.ntnu.no  Sat Aug 28 13:50:04 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 28 Aug 1999 19:50:04 +0200
Subject: Fuzzy string matching?
References: <Pine.LNX.4.10.9908281253390.9521-100000@starship.python.net> <y0jpv07u80k.fsf@vier.idi.ntnu.no>
Message-ID: <y0jlnavu75f.fsf@vier.idi.ntnu.no>

Yet another version, without the append...

def distance(a,b):
    "Calculates the Levenshtein distance between a and b."
    n, m = len(a), len(b)
    if n > m:
        # Make sure n <= m, to use O(min(n,m)) space
        a,b = b,a
        n,m = m,n
        
    current = range(n+1)
    for i in range(1,m+1):
        previous, current = current, [i]+[0]*m
        for j in range(1,n+1):
            add, delete = previous[j]+1, current[j-1]+1
            change = previous[j-1]
            if a[j-1] != b[i-1]:
                change = change + 1
            current[j] = min(add, delete, change)
            
    return current[n]


I *could* have written [i]*(m+1) - but I thought this was clearer... :)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From vitus at wagner.rinet.ru  Tue Aug 31 14:36:22 1999
From: vitus at wagner.rinet.ru (Victor Wagner)
Date: 31 Aug 1999 22:36:22 +0400
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <935457373snz@vision25.demon.co.uk>
Message-ID: <7qh7b6$g9g$1@wagner.wagner.home>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
: In article <7ps4ma$n06$1 at wagner.wagner.home>
:            vitus at wagner.rinet.ru "Victor Wagner" writes:

:> 
:> This is tcl/tk. Almost no difference.

: If that is what you think, you obviously have a different aesthetic
: sense than me.

Of course, I have. I think that application is intended to do the job
not to draw funny pictures on screen, unless it is screensaver.

: Don't forget that Parrot will be able to do more than what is in that
: simple example. For example:

:    sub.fcol=red

But in tcl you can do

option add mywindow*Button.Background red
and achiveve the same effect. Moreover, in Parrot it seems to be coded
into app, but with Tcl option command it can be overriden by user
Xdefaults file.

: would mean ``make all components inside the current component
: default to having an attribute ``fcol'' with thew value ``red'' 
: (i.e red foreground); these defaults can be overridden by the
: attributes in the actual components''

: Creating a default attribute for a component-type will also
: be possible, e.g:

:    all.label.font=Helvetica

It is also possible with option database. 

: Could it write C output that calls the GTK+ library, or C++/Qt,
: or other language/GUI-toolkit combinations?
:  
In principle could. Just define set of procedures named button, label
and so on which generate C source and use them instead of Tk.

But I think that hard coding interface into C code is not a good
approach. You should divide functionality of your application into small
pieces which are provided as scripting language commands. Thus you
clearly separate functionality from interface and if then you decide to
make web-based interface or command-line only interface for batch jobs
which run nightly, it would be simple.


: Would anyone want to do this in Tcl as opposed to Python? (Not
: wishing to get into another language war :-)).

I'm saying for scripting languages as a concept. I just happen to like
Tcl more than Python or Perl or Scheme, but them all are providing
simular levels of abstraction.

:> And parrot files would become runnable. But pretty
:> useless, becouse I've not seen any way to attach functionality to these
:> windows.

: Use subclassing.
No, thanks. I've large experience of programming with toolkits where you
need to create a subclass for each of hundreds dialogs in your program.
I'm avoiding OOP since that as much as possible. Of course, there are
cases, where OOP is The Right Thing, but interface design is not one of
them. At least if you need to subclass for each bit of functinality.
-- 
--------------------------------------------------------
Victor Wagner @ home       =         vitus at wagner.rinet.ru 
I don't answer questions by private E-Mail from this address.



From abigail at delanet.com  Sun Aug 15 05:47:55 1999
From: abigail at delanet.com (Abigail)
Date: 15 Aug 1999 04:47:55 -0500
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ce93@cs.colorado.edu> <slrn7r9la9.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nde.h3c.jstevens@bamboo.verinet.com>
Message-ID: <slrn7rd389.a5.abigail@alexandra.delanet.com>

John Stevens (jstevens at bamboo.verinet.com) wrote on MMCLXXIV September
MCMXCIII in <URL:news:slrn7r9nde.h3c.jstevens at bamboo.verinet.com>:
%% On 14 Aug 1999 02:32:12 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
%% >On 13 Aug 1999 20:04:03 -0700, John W. Stevens <jstevens at basho.fc.hp.com> wrote:
%% >>> In comp.lang.perl.misc, 
%% >>>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
%% >>> :$b[ 2 ] = $c;
%% >>> :
%% >>> :> That's just fine in Perl.  It's not fine in Python, because Python
%% >>> :> won't automatically grow an array.
%% >>> :
%% >>> :'Cause it doesn't have arrays (or, at least, not built in ones).
%% >>> 
%% >>> Gosh, that's a feature.  NOT.
%% >>
%% >>Perl doesn't have lists.  Python doesn't have built-in arrays.
%% >
%% >You should learn some perl you now..
%% >
%% >@array = (1,10,20,30);
%% >$from_list = (1,10,20,30);
%% >$from_array = @array;
%% >print "$from_list\n$from_array\n";
%% >
%% >Will output :
%% >30
%% >4
%% 
%% The @ prefix denotes an array.  You, yourself, should learn
%% Perl.  Calling an array a list, doesn't make it one.

He doesn't. Please, get yourself a copy of Learning Perl and learn Perl.
I don't see any @ in "(1, 10, 20, 30)", so your point is totally moot.

%% >Perl has lists,
%% 
%% Not built in, it doesn't, unless you define array and list as being
%% different words for exactly the same type/class.

Arrays and list are 2 different things in Perl. Perl has lists. They
are however, not the same as Python lists. Don't assume that because
someone mentions "list", Python lists are meant.

%% >if you know perl you would know this.
%% 
%% I know Perl.  You need to learn Python.

The extend of Perl knowledge you have shown in this thread is less
than the average script kiddie knows.

%% >If you program in perl 
%% >and don't know this, then you must get very very confused at times.
%% 
%% If @ denotes list, then the following Perl would be illegal:

But it doesn't denote lists. Your premise is wrong.

%% @ary = (1, 2, 3);
%% @ary[5] = "Test";
%% 
%% But, obviously, this is not illegal.

Indeed. The second line is considered bad style though, and it will
be flagged as a warning with -w. What the second line has to do with
lists, I have no idea. The first line contains a list, on the RHS of
the assignment.

%% >>I will assume that a list module is available for Perl.
%% >
%% >No it is one of the built in bits... like hashes and arrays.
%% 
%% Really?  What is the prefix character that denotes a list?

There isn't. Just like there isn't a prefix character for operators.
There isn't a prefix character for decimal numbers either. What's
your point?

%% >>I wasn't trying to compare features, I was simply pointing out
%% >>that your comparison was Apples and Oranges, and therefore at
%% >>least somewhat invalid.
%% >
%% >Only because you have no idea what you are talking about.
%% 
%% :-)
%% 
%% Coming from somebody who doesn't know the difference from *EMULATING*
%% a list with an array, vs. a real array, that is a good one!

Let me repeat that: You have no idea what you're talking about.

%% Now I suppose that you will tell me that Perl has stacks, too!

I don't think he will. Sam has shown more Perl knowledge than you have.



Abigail
-- 
package Just_another_Perl_Hacker; sub print {($_=$_[0])=~ s/_/ /g;
                                      print } sub __PACKAGE__ { &
                                      print (     __PACKAGE__)} &
                                                  __PACKAGE__
                                            (                )


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From fredrik at pythonware.com  Wed Aug 18 05:26:18 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed, 18 Aug 1999 09:26:18 GMT
Subject: ANN: PythonWare Sound Toolkit (PST) release 0.1
References: <007d01bee8b7$6ecbc5e0$f29b12c2@secret.pythonware.com> <37B983C6.48E6BEAD@hks.com>
Message-ID: <003d01bee95b$bb71dd90$f29b12c2@secret.pythonware.com>

Pedro Proenca <proenca at hks.com> wrote:
> How come it doesn't support MP3?

do you mean "why doesn't the Windows soundplayer
code that deals with MP3 work?" (we're working on that).

or do you have MP3's that are not properly identified by
the MpegSoundPlugin?

or did you just miss that it's an early development release?
(as implied by the version number and the release notes)

</F>




From tchrist at jhereg.perl.com  Thu Aug 19 17:55:47 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Thu, 19 Aug 1999 15:55:47 -0600
Subject: No man pages? (was Re: More random python observations from a perl programmer) 
In-Reply-To: John Landahl's message of Thu, 19 Aug 1999 14:47:34 -0700 .
             <19990819144734.C28466@digex.net> 
Message-ID: <199908192155.PAA08720@jhereg.perl.com>

>If your problem was mostly with 1) then I concur, and so would most Unix
>types, I think.  The rest of this debate seems centered around issue 2), and
>I'm not sure that was your main point.  I agree that 1) should be the
>minimum, but I would tend to look for web-based cross-referenced
>hypertextified documentation for more detailed information.  Though I've
>definitely found "man perlfunc" to be extremely useful when I want to get at
>something quick, so I agree that Python manpages would be a nice addition.

% (echo 'import os' ; echo 'print os.execv.__doc__' ) | python 
Execute an executable path with arguments, replacing current process.

        path: path of executable file
        args: tuple or list of strings

Well, that's better than nothing.

% (echo 'import os' ; echo 'print os.__doc__' ) | python 
None

But that isn't. :-)

I'd guess for a module, you should traverse its members and
print docs for all that have them.

I'm still getting ideas for autogenerating manpages. :-)

But I haven't looked at the HTML or latex docs yet.  Probably
won't till after the conference or so.

--tom



From mwh21 at cam.ac.uk  Thu Aug 19 18:31:08 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 19 Aug 1999 23:31:08 +0100
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu>
Message-ID: <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk>

I sooo want to keep out of this; just a couple of points.

Tom Christiansen <tchrist at mox.perl.com> writes:
> :> GOTCHA: (medium)
> :>     Perl's hex() function is the opposite of Python's.  Perl's hex
> :>     function converts 22 into 37.  Python's hex function converts 37 into
> :>     0x22.  Oh my!  Like hex, Python has oct() completely backwards from
> :>     Perl.  Python's oct(44) returns 054, but Perl's oct(44) return 36.
> :
> :This would have to do with strings not being automatically coerced
> :into numbers, methinks.
> 
> No, that's not it.  Perl's hex takes a hex number and returns 
> a decimal one; same with oct.  Python's is the reverse.  Very weird.
> 
>     echo 'print oct(12)' | python
>     14
> 
>     echo 'print oct(12)' | perl -l
>     10
> 
> See what I mean?

Well, in Python an integer's an integer. It has no property of being
hex or decimal or octal. So by the time the hex/oct function sees it,
it's just an integer irrespective of the way it was typed.

It would seem to me that perl's

oct(12)

is like Python's 

012

A difference.

> :There's info format if you look for it and you have to have a CLI.
> 
> At the very least, python needs a top-level python(1) manpage that gives
> calling conventions, options, envariables etc, and an explanation of how
> to locate the rest of the documentation (even if it's in non-man form).
> That's a POSIX.2 requirement.

On this subject I'll just say: If people before you had cared as much
about the existence of a python(1) manpage, it would have one.

> :> GOTCHA: (low)
> :>     Often Python's error messages leave something to be desired.
> :>     I don't know whether 
> :
> :Is this a joke? This is improving.
> 
> No, it isn't a joke.

I was referring to the unfinished "I don't know whether" line.

[explanantion of slicing dicts elided]

Oh, I see.

> :new = old.copy()
> 
> I wonder how many times I'll have to explain that I read the 
> Lutz book, wasn't on the Internet at the time, and found no manpages
> installed with python on my laptop.

Ah, have you discovered rlcompleter yet? Put this in your ~/.pythonrc:

import rlcompleter
readline.parse_and_bind('tab: complete')

set PYTHONSTARTUP=~/.pythonrc and then in python type:

d={}
d.

and whack tab a couple of times. This should definitely be better
advertised; I use it all the time.

> :Yes this is a bit annoying. The Right Way to attach state to a
> :function is a class, of course. Not that that helps much.
> 
> That sounds like religion.

I suppose, but not Python's particularly.

>  And aren't bound methods inherent
> memory leaks?
> 
>     fn = obj.method;

No.
 
> :> GOTCHA: (low)
> :>     Python's eval() only works on expressions, not full code blocks full
> :>     of statements.  You need exec() for the whole thing.
> :
> :Writing exec() suggests it's a function; it isn't, it's a statement.
> 
> Funny that you guys are all flustered about this statement/function thing.
> I think of if and def as statements.  I'm not saying print and exec
> aren't, but they sure look just like it.

I was just being accurate.

Cheers,
Michael



From jbauer at rubic.com  Mon Aug 16 17:58:28 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Mon, 16 Aug 1999 21:58:28 GMT
Subject: Python CGI Scripts
References: <Pine.LNX.4.04.9908162019270.3026-100000@natal.stamp-coin.com>
Message-ID: <37B88984.75693541@rubic.com>

> From my POV, Zope has two major problems.
> 
> #1:     It expects to run as a server.  This is not a good
>         thing, from the POV of most IPPs and ISPs.

Zope usually runs as a server (a.k.a. long running process)
because it is more responsive that way.  It can also run in a
conventional CGI manner (one new process forked per hit)
if you have low traffic and/or don't mind paying the price.

For many low-traffic tasks, I don't use an LRP.  My code
uses the publisher (maps urls to methods & paramenters),
document templates, and the persistent object store (POS).
The latter is basically a pickling mechanism.  You write
Python class methods which are invoked as http requests.

Also:  Although Zope is usually run as a server, this
does not mean it expects to be _the_ server.  Your ISP
can use their favorite web server (e.g. Apache) to 
communicate with a Zope process.

This begs the question: Will your ISP allow you to own
a long-running process on one of their machines?  Many
won't. but some ISP's explicitly provide web hosting 
services to Zope users.

Best regards,

Jeff Bauer
Rubicon, Inc.



From claird at starbase.neosoft.com  Sun Aug 22 23:27:19 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 23 Aug 1999 03:27:19 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu>
Message-ID: <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net>

In article <37b634a2 at cs.colorado.edu>,
Tom Christiansen  <tchrist at mox.perl.com> wrote:
			.
			.
			.
>Think back to the 80s and early 90s.  The change, if it really happened,
>came with the upty-gazillion CGI script kiddies who couldn't program
>2+2, yet who wanted *us* to write *them* these persistent, encrypted,
>crossplatform, multiscreen shopping carts for e-commerce, complete with
>dynamically-generated animated vanity counters, each and every time 
>they stepped up to the feeder.
			.
			.
			.
And e-mail address validation; don't forget that.
No, wait, I mean, the *eight* things no one ever
expects from the Spanish Inquisition are ...
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From drew at getaway.net  Sun Aug 15 04:09:14 1999
From: drew at getaway.net (Andrew McDowell)
Date: Sun, 15 Aug 1999 03:09:14 -0500
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <000501bee6cb$a31710e0$832d2399@tim>
Message-ID: <37B675A9.53C2BBE0@getaway.net>

First off, I'd like to apologize for the Compound Assignment Operator vs.
Compound Assignment Statement confusion...
I'll get the hang of this some day <grin>


Tim Peters wrote:

>  Seriously, the problem at the start was simply one of
> non-obviousness.  For example, if x and y are lists, does
>
>     z = x
>     x += y
>     print z
>
> mean
>
>     x = x + y   # x is now bound to an entirely different object
>     # and the old x is printed
>
> or
>
>    x.extend(y)   # x is still the same object, but mutated
>    # and the new x is printed
>
> ?  The resulting *values* compare equal either way, but there are
> differences in efficiency and effects on object identity.
>

I don't see where the "non-obviousness" comes into play...
(maybe that's why it's not obvious <wink>)

Why would you confuse x += y (which would seem to imply the former) to mean
x.extend(y) ?   It _is_ and assignment statement after all..


>
> There were more important questions to resolve first, and the lack of
> augmented assignments hasn't caused enough people enough pain to force a
> resolution of these questions.

Not to touch on a sore topic, but hop over to comp.lang.perl and check out the
(rather long and heated) comparison of perl and python.
The lack of these statements is causing quite a bit of a stir.


>  Nevertheless, in good Pythonic fashion,
> Guido's favored resolution is to say it's up to the object:  whatever type x
> may happen to have, it's up to x.__addeq__(y) to decide what "+=" (& so on)
> means.
>

Why addeq??  Why not just add?

>
> 5) Not having it was good enough for my grandfather, and what makes you
> think you're better than him <wink>?
>

Because Python wasn't around then! <wink>






From guy_oliver at yahoo.com  Fri Aug 27 07:29:18 1999
From: guy_oliver at yahoo.com (Guy Oliver)
Date: Fri, 27 Aug 1999 11:29:18 GMT
Subject: Why do we call python a scripting language?
Message-ID: <19990827112918.12652.rocketmail@web129.yahoomail.com>

> <guy_oliver at yahoo.com> wrote:
> > Why do we call python a scripting language?  
> 
> who's calling it that? ;-)
> 
> http://www.python.org/doc/Summary.html
> says:
> 
>     Python is an interpreted, interactive, object-
>     oriented programming language.
> 
> which is, of course, exactly what it is.
> 

I don't think I've seen anyone 'official' (meaning guido et al) say
anything like that, but I see it used in the media (must have come
from someone), in news postings, and in people trying to explain
what python is to others.

Now, I know there is no "one true language", or if there is, I
haven't figured out what it is.  And I am not trying to say that
python is or is not a "real" programming language (what ever that
is), but I have often tried to pitch python to someone as a
solution to a problem, expecting some resistance due to
unfamiliarity, only to have someone ask if its a "real programming
language".  

Even after explaining it to them, and telling them my success
stories, the improvements in productivity, the portability, etc, it
seems that python suffers from the same view that hurts other
interpreted languages, specifically, that interpreted languages,
aka scripting languages, are for throw-away scripts, nothing more.

I'm wondering if there is anything that can be done to combat this?
 With education, it goes away, but I expect people to equate script
with simple for a long time, so the quickest solution seems to be
try to distance 'python' from 'script'.




__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com




From dfan at harmonixmusic.com  Tue Aug 17 16:44:28 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 17 Aug 1999 16:44:28 -0400
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com>
Message-ID: <wklnbagms3.fsf@turangalila.harmonixmusic.com>

Ole-Kristian Villaboe <olevi at my-deja.com> writes:

| Is there a command to do this? (Can't find one but...)
| I've been writing an algorithm myself, but it's incredibly slow and I
| would like something quicker (of course :) )

This is slower and takes more space than the correct solution already
posted, but it's kind of fun:

def randomize (l):
    l2 = map (lambda x: (whrandom.random(), x), l)
    l2.sort()
    return map (lambda x: x[1], l2)

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From aahz at netcom.com  Sun Aug 15 17:54:20 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 15 Aug 1999 21:54:20 GMT
Subject: C's isprint() concept?
References: <slrn7rea88.9uf.jblaine@shell2.shore.net>
Message-ID: <7p7cuc$plr@dfw-ixnews5.ix.netcom.com>

In article <slrn7rea88.9uf.jblaine at shell2.shore.net>,
Jeff Blaine <jblaine at shell2.shore.net> wrote:
>
>If I want to replace all non-printable characters in a string with a single
>space, what would be the best way?  Do I need to loop over the entire string
>character by character checking the ord() value of each one?  Anyone have
>a sane way to do this with regular expressions?

Oh, sure.  What isn't clear from the way you write this is whether a run
of multiple non-printable characters should be replace with a single
space.  Here's a regex that I created recently as part of some code to
detect binary documents:

   # Everything that isn't CR/LF or 7-bit normal characters
   reBinary = re.compile ( r'[^\r\n\x20-\x7F]' )

To make this substitute a space for each character, just do

   newString = reBinary.sub ( ' ', string )

I leave the multi-sub as an exercise for the reader.

>[ C has isprint() and isgraph() macros ]

Don't remember offhand what isgraph() does, but it occurs to me that it
would be useful to have a character class string.printable.  Tim?  Guido?
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From hniksic at srce.hr  Tue Aug 31 20:12:38 1999
From: hniksic at srce.hr (Hrvoje Niksic)
Date: 01 Sep 1999 02:12:38 +0200
Subject: Python speed and `pcre'
References: <oqzoz89hrb.fsf@titan.progiciels-bpi.ca>
Message-ID: <87lnara3rd.fsf@pc-hrvoje.srce.hr>

Fran?ois Pinard <pinard at iro.umontreal.ca> writes:

> After having translated some code (not big, but not small) from Perl
> to Python, I discover it runs ten times slower.

This is my usual result when comparing just about any code where speed 
actually matters (for me it's usually processing of large textual
files.)  I don't think the speed different is related to compiling the 
regexps, although it is certainly possible.  Python is significantly
slower at executing regexps, as well as in simple tasks such as
reading from a file.

In most cases this is not a problem, but sometimes, like with the said 
large files, it is really disappointing to think about rewriting a
very nice piece of Python code to Perl for efficiency reasons.

People who embrace Python as a "better Perl" should really beware the
speed difference.



From mwh21 at cam.ac.uk  Fri Aug 20 05:39:12 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: Fri, 20 Aug 1999 10:39:12 +0100
Subject: More random python observations from a perl programmer
In-Reply-To: <37BD1A8A.D4DCFC7C@strs.co.uk>
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu> <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk> <37BD1A8A.D4DCFC7C@strs.co.uk>
Message-ID: <Pine.LNX.4.10.9908201036330.21136-100000@atrus.jesus.cam.ac.uk>

On Fri, 20 Aug 1999, Ian Clarke wrote:

> 
> > Ah, have you discovered rlcompleter yet? Put this in your ~/.pythonrc:
> > 
> > import rlcompleter
> > readline.parse_and_bind('tab: complete')
> > 
> > set PYTHONSTARTUP=~/.pythonrc and then in python type:
> > 
> > d={}
> 
> Is there any way to get this working on Windoze?

Not that I know of; there is a port of readline to Windows (or at least I
found a reference to one last time I looked) but I don't know how much
work would be involved in getting Python to play nice with it.

Or Pythonwin could be modified to include such a feature. Now idea how
much work that would be either. 

(Tab completion (nad I guess readline editing features in general) is the
main reason I still use the interpreter over idle.)

HTH,
Michael




From tim_one at email.msn.com  Sun Aug 29 17:32:27 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 29 Aug 1999 17:32:27 -0400
Subject: Returning none
In-Reply-To: <slrn7sgrse.4ee.clgonsal@keeshah.penguinpowered.com>
Message-ID: <000001bef265$fe64f340$192d153f@tim>

[C.Laurence Gonsalves]
>|> ...
>|> Personally, I think falling out of functions and expecting
>|> None to be implicitly returned is a very Perl-ish thing to do...

[Tim]
>| [sez Perl actually returns the last expression evaluated, which
>|  is what CLG intended in his example]

[CLG]
> Yes, in this case. I was referring to the way that Perl tries
> to guess what you meant rather than treat dubious code as
> incorrect.

Except that Python isn't trying to guess what you mean here -- as before,
functions returning None by default is not a convenience, it's a cheap way to
make an error likely when mistakenly using a procedure as a function.  It's not
a bulletproof approach, but it is *an* approach, and mischaracterizing its
nature weakens your case.

>|> Why don't we add $_ to Python as well?

>| [and why $_ makes more sense in Perl than it would in Python]

> I was being sarcastic.

I know -- that's why I wasn't <wink>.  Dismissing ideas out of hand as all loss
and no gain is as unfruitful as embracing suggestions as if all gain and no
cost.  As PaulP said, in reality it's a balancing act.

> I was pointing out that if Python is going to start making wild
> guesses about one thing, then it might as well do it everywhere
> else as well.

Except (a) it's not making wild guesses here; and, (b) compromises are indeed
made everywhere anyway, so we're already at the bottom of the slippery slope
you fear <wink>.

>| ...
>} Python doesn't return None for [convenience]; it's instead a cheap
>| way to make [an error likely]

> That sort of argument

It's not an argument:  it's the fact of how this particular compromise came
into being.

> also says that when I go outside the bounds of a list in Python,
> I should get back None, since in C I would access random stack/heap
> trash.

I agree that having an out-of-bounds list reference silently return None would
also be a way to make it (merely) likely to lead to an error.

> Instead, Python raises an exception. I think it should do the same
> if you don't return anything, and then try and access that nothing.

If there were no costs associated with this, or if the costs were thought
smaller than the benefits, don't you think Python already would?

In the list case, it cost nothing:  no additional syntax or long-winded rules
were required, and the interpreter had to check for out-of-bounds anyway (else
risk having the interpreter itself blow up due to a bogus C operation).

As to benefits, I'm with most of the other respondents:  I can't recall this
insecurity (it certainly is one!) ever causing a significant problem -- and
I've been programming in Python even longer than Tom Christiansen <wink>.

> ...
> That's why I also suggested some other possible rules that would
> enable some amount of compile-time checking, so that function
> writers who do questionable things would be alerted.

The best hope for this is that it gets folded into the push for optional type
declarations in Python2; that will need a way to indicate function return types
(or lack thereof) anyway.

...

>| Do you imagine Guido didn't consider this when he designed
>| the language?

> So Python is *absolutely perfect* and shouldn't be changed at all?

No.  You showed no sign of having considered why it might be the way it is, so
I was inviting you to do so.  Sarcasm was not the hoped-for result.

> I guess us mere mortals shouldn't even bother making any suggestions
> then.

Ditto.

  Python 2 will just be Python 1.5.2 with a new name...

Ditto.

> In another post I suggested another idea that requires little or no
> modifications to the grammar, but would just add some new runtime and
> compile time checks.

You should study eval_code2 (in Python/ceval.c) if you want to suggest
implementations.  That would give you a handle on part of the "cost" side of
the tradeoffs.

> Functions using return (or falling off the end) would return nothing
> at all (rather than None),

The interpreter can't do that as currently structured -- a call must return a
pointer to a PyObject.  You could invent a new internal NotARealObject type and
return a pointer to an object of that type, but then it gets messier.

Currently a CALL_FUNCTION opcode simply pushes the object it gets back -- it
has *no idea* whether the result will be used later or discarded, so doesn't
have enough info to know whether to complain.  So you'll have to make
CALL_FUNCTION very much smarter about its context.

But then you have miserable semantic issues to resolve; for example, while you
clearly must allow f to return nothing given the statement

    f()

is it also OK for f() to return nothing in the statement

    f(), g(), h()

?  You must allow that else break existing code, but the result of f() *is*
used by a later BUILD_TUPLE opcode:  it's not the case that "returning nothing"
can work here, even if the interpreter were structured in a way that such a
thing were possible -- BUILD_TUPLE requires an object.

In this example, the tuple itself is later discarded, but there's no way for
CALL_FUNCTION to know that without prior flow & lifetime analysis (Python has
neither today), and a rigorous definition of which opcodes do and don't count
as "real uses".

Simply returning None was a pragmatic compromise that saved a world of
implementation pain in return for a tiny chance of letting an error go
undiagnosed forever, and an excellent chance of catching an error but later
than you might hope for.

> and function definitions would require that the function either
> always return a value, or never returns a value. This can be very
> easily checked at compile time.

While this part is more clearly doable, it's also a probabilistic compromise.
In your original example, it wouldn't complain (you had no returns, so this
scheme would let it pass).  Better for optional declarations to allow you to
say explicitly whether you intend to return a result or not, rather than
replace one "guessing" scheme with another.

It also introduces an ugliness Python doesn't suffer today; e.g.,

def first_factor(n):
    for i in xrange(2, n):
        if n % i == 0:
            return i
    complain(ValueError, n, "doesn't have a non-trivial factor")

def complain(exc, *args):
    msg = string.join(map(str, args), " ")
    raise exc(msg)

There's nothing wrong with either function, but to stop a bogus complaint the
first would need a

    return 0   # make the stupid compiler shut up

line at the end.  To avoid this kind of artificial ugliness in a language with
exceptions requires something akin to Java's (many) pages of "definite
assignment" rules.  I happen to *like* Java's rules here (they catch many
errors at compile-time Python cannot, and almost never give me bogus
complaints), but without static declarations (or global inference coupled with
weakened dynamics) Python can't approach that.

nothing-is-a-pure-win-ly y'rs  - tim





From fredrik at pythonware.com  Thu Aug 19 03:30:06 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 19 Aug 1999 07:30:06 GMT
Subject: Unsorting(randomizing) a sequence
References: <000501bee917$d55fd660$eaa22299@tim>
Message-ID: <005201beea1a$5ab05f30$f29b12c2@secret.pythonware.com>

Tim Peters <tim_one at email.msn.com> wrote:
> Python implements its own sort function (in 1.5.2 it's not quicksort, btw,
> but a hybrid of common special cases, binary insertion and samplesort).  A
> very long time ago it used the platform qsort, but that blew up on some
> systems; iirc, usually due to qsort being implemented non-reentrantly.  The
> current sort is robust against all that kind of stuff, including insane
> comparison functions; it always returns *some* permutation of its input.

how about JPython and Viper and Lumberjack
and SPython and FooPython and whatever...

as far as I can tell, the "spec" says 

    sort: sort the items of s in place

and that's about it...

</F>




From thomas at xs4all.nl  Fri Aug 20 11:53:47 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Fri, 20 Aug 1999 17:53:47 +0200
Subject: Setting environment variables.
In-Reply-To: <37BD3A33.6CAA145A@fnmail.com>
References: <37BD3429.7364B4F7@iinet.net.au> <37BD3A33.6CAA145A@fnmail.com>
Message-ID: <19990820175347.F461@xs4all.nl>

On Fri, Aug 20, 1999 at 01:21:23PM +0200, Per Kistler wrote:

> Oliver White wrote:
>  eg, after
> > running my python program as a child process, will the changes to the
> > environment variable affect it's parent (the *term)? (suspecting no) Is
> > there annother way to set the parent's?

> I'm wondering, whether one could exec to a program which sets
> ones env and then exec back to a shell, which would then have
> the desired env. One would no more be in the original shell,
> but the new one would have the desired env.
> Maybe some guru could elaborate on that....

It gets better... C's exec*() family allow you to specify the environment.
And so does python :)

(from the Python Library Reference 6.1.5, os module, Process Management)

execle (path, arg0, arg1, ..., env)
      This is equivalent to "execve(path, (arg0, arg1, ...), env)".
      Availability: Unix, Windows.

[..]

execve (path, args, env) 
      Execute the executable path with argument list args, and environment
      env, replacing the current process (i.e., the Python interpreter). The
      argument list may be a tuple or list of strings. The environment must
      be a dictionary mapping strings to strings. Availability: Unix,
      Windows

>>> x = UserDict.UserDict(os.environ)
>>> x["spam"] = "eggs"                          

Just so you know i aint cheating this time ;-)

>>> os.environ["spam"]
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python1.5/UserDict.py", line 14, in __getitem__
    def __getitem__(self, key): return self.data[key]
KeyError: spam

>>> os.execle("/bin/tcsh", "-", x)
centurion:~/python > echo $spam
eggs

Yes, you could set os.environ before an os.exec*(), but passing a modified
env is nicer.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From skip at mojam.com  Fri Aug 27 23:39:15 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 27 Aug 1999 22:39:15 -0500 (CDT)
Subject: RFI: What changes in Python 1.6
In-Reply-To: <37c72dee.1411808533@news.triode.net.au>
References: <37c72dee.1411808533@news.triode.net.au>
Message-ID: <14279.22557.427799.456090@dolphin.mojam.com>

    John> Would someone with access to Python 1.6 working sources like to
    John> summarise changes from Python 1.5.2 for me and others that might
    John> be interested?

You can grab the latest source from the CVS repository.  Check the Python
website.  Functionally, I believe string objects will get methods, mostly
dispensing with the need for a separate string module.  Thanks to Barry
Warsaw, that code is in the CVS tree now.  You can experiment with them if
you pull out the string_methods branch of the tree.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From kistler at fnmail.com  Wed Aug 18 05:19:28 1999
From: kistler at fnmail.com (Per Kistler)
Date: Wed, 18 Aug 1999 11:19:28 +0200
Subject: Wacky Programming Tales (Was Re: Why use Perl when we've got 
 Python?!)
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <7pd7f9$ais$1@ssauraaa-i-1.production.compuserve.com>
Message-ID: <37BA7AA0.1C67A7A6@fnmail.com>

Such things are still harmless compared to the macros of the
C++ compilier of HP-UX. They put a variation of the standard template
library into macros (,in certain include files).
They argue, it's not ansi c++, which does not help me that much,
and neither reduces the price of the compiler. (To say something
nice about HP, the 8cpu machine runs quite nicely:-).

And a small positive statement about Perl and Python: Both are
not Tcl:-) (, and no one from comp.lang.tcl can read this:-)
And I like both, Perl and Python. Perl and Tcl brought me some
money already and Python will sure do so in the future as well. 
If one collects the advantages of all three scripting languages, 
one can do nearly everything.

-Per.

"T. C. Mits" wrote:
> ...
> you don't use a good lint, because the boss won't buy one), #define EQ ==,
> is not a sin.   Well, yes it is, but its not, but.....

> >   #define AND &&
> >   #define OR  ||


-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From pacman at defiant.cqc.com  Thu Aug 19 15:34:21 1999
From: pacman at defiant.cqc.com (Alan Curry)
Date: Thu, 19 Aug 1999 19:34:21 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <7phfe0$19so$1@nntp6.u.washington.edu> <37bc487e@cs.colorado.edu> <7phk5v$ot1@dfw-ixnews8.ix.netcom.com>
Message-ID: <1%Yu3.11298$x04.661068@typ11.nn.bcandid.com>

In article <7phk5v$ot1 at dfw-ixnews8.ix.netcom.com>,
Aahz Maruch <aahz at netcom.com> wrote:
>I can't tell you how many times I've gotten pissed at having one Perl
>man page reference another with no direct mechanism for getting there.

I bet it doesn't outnumber the times I've gotten pissed at lynx for not
having a regexp search.

-- 
Alan "If it doesn't go through my $PAGER, it's not documentation" Curry



From thomas.heller at ion-tof.com  Tue Aug  3 11:27:23 1999
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Tue, 3 Aug 1999 17:27:23 +0200
Subject: Getting "Short" filenames inWindows
References: <C2AA61E5CAA7D211832F0008C7A4138E01B4873C@eurmsgneu01.eur.autodesk.com>
Message-ID: <00a801beddc4$af575fa0$4500a8c0@thomasnotebook>

This was answered some time ago by Tim Peters.
Read the full story at:
http://x39.deja.com/[ST_rn=ps]/getdoc.xp?AN=475004666&CONTEXT=933693577.1655
177350&hitnum=0

This URL was found by searching for 'short filenames convert' in the
'Search the Python newsgroups (using DejaNews)', which you will
find on http://www.python.org/search/.

Regards

Thomas Heller
ION-TOF GmbH
----- Original Message -----
From: <mark.mcmahon at eur.autodesk.com>
To: <python-list at python.org>
Sent: Tuesday, August 03, 1999 1:32 PM
Subject: Getting "Short" filenames inWindows


> Hi Everyone,
>
> Is there some simple way to get the short path name
> from the long one in Windows.
>
> os.path, dospath and ntpath files do not seem to
> allow one to access the short name easily.
>
> Thanks
> Mark
>
> --
> http://www.python.org/mailman/listinfo/python-list
>




From conversy at lri.fr  Tue Aug 24 11:27:11 1999
From: conversy at lri.fr (Stephane Conversy)
Date: Tue, 24 Aug 1999 17:27:11 +0200
Subject: *** Parrot-0.0.1 released ***
References: <935279776snz@vision25.demon.co.uk> <935454399snz@vision25.demon.co.uk>
Message-ID: <37C2B9CE.95AA5FEC@lri.fr>

Could you give an example of how to plug code with a generated python
file ?

for example, how to attach a command when the "New" menu is triggerred ?

window @MyWindow "My First Window" {
     menuBar {
        menu "File" {
           menuItem @New "New"
        }
}

    stef

--
St?phane Conversy
http://www-ihm.lri.fr/~conversy/
mailto:conversy at lri.fr






From alan.gauld at gssec.bt.co.uk  Fri Aug 27 12:27:36 1999
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Fri, 27 Aug 1999 17:27:36 +0100
Subject: UML
References: <37B96A3A.796BB034@compuserve.com> <mmNw3.12$z3.2904@newsfeed.avtel.net>
Message-ID: <37C6BC78.95490313@gssec.bt.co.uk>

Phil Mayes wrote:
> robert_wilhelm_land wrote:
> >Are any persons in this group using the unified modelling language?

Yes.

> >If yes, could you kindly explain when to use the UML and what the
> >advantages are?

Its a standard notation that means if several 
people/organisations are working on the same system 
you can all share/understand each others design models. 
If you are working in Python this is not too likely 
to apply!

UML is also usefiull in modelling business processes and 
hardware deployment designs (which are the areas I 
currently use it)

<snip a rather harsh description of the 3 amigos :-) >
> their skill level, consider this: they chose to draw their class
> inheritance trees with the arrows going up towards the root.

Which is sensible since arrows in design diagrams 
often represent dependency relationships and the 
derived class *depends on* the superclass.

Also in an inheritance heirarchy the message processing 
flows up the inheritance tree, starting at the concrete 
class and then up to the parent etc... So even if you take 
the arrows as representing process flow, upwards makes sense.

> It may be a useful job skill when interviewing at a large corporation,
> but IMHO, a whiteboard and 60 minutes thinking wins hands down.

If the system is big or the team dispersed a good notation 
and method are essential.

Whether UML is good is another matter. See my web page 
on that topic :-)

http://members.xoom.com/alan_gauld/uml.htm

Alan G.

-- 
=================================================
This post represents the views of the author 
and does not necessarily accurately represent 
the views of BT.



From da at ski.org  Mon Aug 30 05:36:01 1999
From: da at ski.org (David Ascher)
Date: 30 Aug 1999 04:36:01 -0500
Subject: README: Python-URL! - the weekly guide to Python resources (Aug 30)
Message-ID: <DF8B0C0689913F6B.4885F950288DC6B3.8132F0A49FEEC5C3@lp.airnews.net>

Highly biased highlights from the O'Reilly conference:

    Tom Christiansen joins Guido and a few other Pythonistas for beer
    and random trivia (does || or && bind tighter in C?). A jovial
    time was had by all, and nothing caught on fire.

    There were more folks looking for Python programmers than Python
    programmers looking for a job!  The future's so bright, we gotta
    wear shades.

    eGroups CTO Scott Hassan says that they pump 30 million emails a
    day with Python without breaking a sweat, and expect to scale up
    to 200-300 million personalized emails per day.  Big numbers for a
    'slow' language!

    When Adam Feuer was asked whether he was having fun, he replied:
    "I'm having a blast -- I've been using Python for six months, and
    I feel like I'm having an affair with the language!"

   http://www.linuxslides.com/
   http://www.vwtapes.com/cgi-local/vwcart.pl?page=/oreilly/99/python.html
   
Having the old fogies out of the way spurred activity on the
newsgroup:

    One old fogie remained: Fredrik Lundh "parses arbitrary python
    expressions" he doesn't trust
        http://x32.deja.com/getdoc.xp?AN=506369193 and Bernhard Herzog
    shows how to sneak in statements
        http://x29.deja.com/getdoc.xp?AN=506541895

    Arpad Kiss provides one minimal example of how COM can
    be useful by retrieving document properties into a Py-
    thon process
	http://x46.deja.com/getdoc.xp?AN=508285498

    David Niergarth "unmangles" a Win* "long name", but
    only with difficulty.  No better solutions appear
	http://x27.deja.com/getdoc.xp?AN=510820469

    Colin J Williams describes Hans Nowak's page of
    "snippets" as "clean and well-organized"
	http://www.hvision.nl/~ivnowa/snippets/

    Karl Putland retrieves a collection through COM
	http://x24.deja.com/getdoc.xp?AN=505637404

    Charles G Waldman carefully illuminates several alternative
    ways to shuffle cards
	http://x38.deja.com/getdoc.xp?AN=513914923
    Nihilist Tim Peters, once again misremembering which is the
    "dismal science", counters that "even the best possible al-
    gorithm ... is terrible"
	http://x37.deja.com/getdoc.xp?AN=514317775
    Other poetry from the same thread:  Mersenne Twister, tears-
    in-the-rain, Chi-squares, /dev/random

    People seem to argue a lot about nothing, specifically:
    The necessesity of pass statements in empty blocks, and whether
    returning None is a trap.  Some of the more useful posts:
        http://www.deja.com/getdoc.xp?AN=516955938
        http://www.deja.com/getdoc.xp?AN=518128657

    More usefully, some etymological truths are laid bare,
    specifically regarding nautical techniques thanks to Fredrik
    Henbjork the fisherman
        http://www.deja.com/getdoc.xp?AN=515457138
    and Tim Peters and Kent Polk argue about who the real mathematical
    geniuses are
        http://www.deja.com/getdoc.xp?AN=518132556

    In a blast from the past, what used to be Ted Nelson's Xanadu is
    bared to all, and some might be surprised to know what languages
    it's written in: 
        http://www.udanax.com/

=========================================================================
Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

    Archive probing trick of the trade:
        http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
  http://www.egroups.com/list/python-url-leads/

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.


-- The Python-URL! Team--
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From Pascal.Nelson at analog.com  Fri Aug 13 13:42:10 1999
From: Pascal.Nelson at analog.com (Pascal Nelson)
Date: Fri, 13 Aug 1999 13:42:10 -0400
Subject: Python tkinter with Tcl/Tk 8.1?
References: <7osg3n$l41$1@news.vanderbilt.edu>
Message-ID: <37B458F2.1691D1EE@analog.com>

foo wrote:
> 
> I could not find any documentation on this in the FAQ or the tkinter pages,
> so I am asking here. I have had to move to Tcl/Tk 8.1 for another software
> package I am using on my computer. I have rebuilt Python 1.5.2 after getting
> the new Tcl/Tk installed and (no great surprise) tkinter does not seem to
> work (I can successfully import Tkinter from the python command line, but
> attempting to run Tkinter._test() crashes python due to access violations
> (reading wild pointers).
> 
> Is it possible to build the tkinter module successfully with Tcl/Tk 8.1?
> 
> Thanks,
> Jonathan

I just experienced the same problem a couple of days ago trying to get
Python 1.52 and TCL/TK 8.1.1 working on a Sun running Solaris. I found
out that there is are two very simple patches that need to be applied to
the source of _tkinter.c and tkappinit.c. These patches were posted a
few months ago by Dieter Maurer to the python mailing list. Archives of
the mailing list can be found at
http://www.python.org/pipermail/python-list/

Just to make it easier, here are the patches:

############################################

--- :_tkinter.c Mon Jan 25 22:39:03 1999
+++ _tkinter.c  Fri May 28 21:18:16 1999
@@ -491,6 +491,10 @@
 
        v->interp = Tcl_CreateInterp();
 
+#if    TKMAJORMINOR >= 8001
+       TclpInitLibraryPath(baseName);
+#      endif /* TKMAJORMINOR */
+
 #if defined(macintosh) && TKMAJORMINOR >= 8000
        /* This seems to be needed since Tk 8.0 */
        ClearMenuBar();

############################################

############################################

--- :tkappinit.c        Tue Dec  2 21:38:38 1997
+++ tkappinit.c Tue Jun  1 21:11:25 1999
@@ -21,12 +21,12 @@
 {
        Tk_Window main;
 
-       main = Tk_MainWindow(interp);
-
        if (Tcl_Init (interp) == TCL_ERROR)
                return TCL_ERROR;
        if (Tk_Init (interp) == TCL_ERROR)
                return TCL_ERROR;
+
+       main = Tk_MainWindow(interp);
 
 #ifdef WITH_MOREBUTTONS
        {
############################################

I applied these patches, and it fixed the problem.

Good luck.

Pascal


-- 
Pascal Nelson
Analog Devices Inc.
Greensboro NC



From jstevens at bamboo.verinet.com  Fri Aug 13 23:13:52 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: 13 Aug 1999 20:13:52 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <37b4a13d@cs.colorado.edu>
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4d0e0@cs.colorado.edu>

In comp.lang.perl.misc, you wrote:
>     [courtesy cc of this posting mailed to cited author]
>
>In comp.lang.perl.misc, 
>    "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
>:Which makes Perl perfect for the lone programmer who comes from a
>:Unix/C/Shell background, and who has no OO training.
>
>That statement is trivially countered by demonstrating the zillions of
>Perl OO modules people, have distributed to parties unknown to them.

Zillions?  Are you sure that this isn't a slight exageration?

After all, if what you say re: 100x's of times more Perl users
than Python, then even if OO were a much less used paradigm on
Perl than it is on Python, you'd still expect to see more.

Most importantly: I am not talking about absolutes, here.

A obsessively quality concious, extremely disciplined, very
experienced assembly language code writer can produce highly
efficient, easily maintained code.

The issue isn't one programmer, or even examples of many, but
statistical comparisons.  The existence of a body of Perl code,
being maintained by a group that has a high turn over rate,
does not invalidate my point.

Compare costs, defect rates, etc. across many such groups,
and then you will have some data to work with.

>I completely disbelieve this "lone programmer" crud, and challenge you
>to defend it.

I've seen it repeatedly.  Somebody wants to do something.  They
drag in Perl.  They write in Perl.  Management gets all hot
about it, decides to support it, then the headaches hit.

>I suspect that you're confusing "bad programmer" with
>"lone programmer".

No, I'm not.  I suspect, however, that you are confusing your
level of ability with "the average Perl programmer".

>A bad programmer will make code that you wouldn't
>want to try to share and reuse.

Agreed.

>A good programmer makes good code.

Also agreed.  The issue isn't the lone programmer!  The issue is
*groups* of programmers: teams, in relationship to an organization.

>Please don't pretend that you can't do that in Perl, or that it hasn't
>been done.

I haven't.  I won't.  The issue isn't what can and cannot be done,
the issue is the margin: which is *better*, on the average?  Which
has lower costs?  Which has lower defect rates.

The fact that you, personally, can write low defect, highly
maintainable Perl code is not the issue: the issue is the
average programmer, in the average organization, who has the
average amount of training.

>That there exist millions of horrible script kiddies with
>CGI written all over them doesn't change any of this.

Actually, yes, it does.  Think about it.

John S.




From gherron at aw.sgi.com  Thu Aug 19 19:02:29 1999
From: gherron at aw.sgi.com (Gary Herron)
Date: Thu, 19 Aug 1999 23:02:29 GMT
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk>
Message-ID: <37BC8D05.142E1DB0@aw.sgi.com>

Stuart Reynolds wrote:
> 
> Does anyone know how to tell if Tk has already instaniated a Toplevel
> window? I want to be able to open several Toplevel windows in any order.
> 

Perhaps the easiest way is to never use the default top level window:

from Tkinter import *
DefaultTop = Tk()
DefaultTop.withdraw()  # Never to be seen 

def makeNewWindow():
   mstr = Toplevel()
   f = Frame(master=mstr)
   ...

You can still do
  DefaultTop.mainloop()
when it time to turn control over to the main event loop

-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101



From I.Clarke at strs.co.uk  Mon Aug 16 05:48:01 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Mon, 16 Aug 1999 10:48:01 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu>
Message-ID: <37B7DE51.B2F2549B@strs.co.uk>

> Unbelievably stupid subject, eh? 

Yes, but there is a subtle difference between this and my original
subject of "Why use Perl when we've got Python?".  One difference
between Python and Perl is that Perl was already around when Python was
first created.  Thus, to suggest that Python is unnecessary given the
fact that we have Perl is saying that Guido wasted his time in creating
Python.  The suggestion that Perl could be superseded by Python (which
is a more recently conceived language) is less ridiculous.

Perhaps my original subject was designed to grab people's attention, but
the body of the message indicated that I was merely seeking knowledge
about Perl from those who know more than me.  If the quality of a post
is to be determined solely by its subject, then you are just as guilty
as you perceive me to be (for posting with this subject).

> I certainly thought so when it was
> shoved at us.  The flame war begun in comp.lang.perl.misc isn't doing
> anyone any good, and then coming over to comp.lang.python to stir up the
> shock troops to send in a few commandos is even worse.  I don't think
> you'd like it if we in comp.lang.perl.misc sent 100,000 script kiddies
> over to comp.lang.python to rant at you guys about how K31L they though
> Perl was and how L at M3 they thought Python was.

Actually of all the replies to my posts, yours were the only to
descended into personal insults, even as I did my best to remain polite
and respectful for someone I don't know.  Particularly amusing that you
(apparently) have killfiled my emails when it was you who started carbon
copying my email address in your replies.

As for the "flame war" doing nobody any good, I think it has answered my
original question quite nicely, and probably educated many about the
pros and cons of Perl versus Python.

Ian.

> Do you want to learn or do you want to flame?  Learn?  Really?  Ok, here's
> all you have to do.  Go write a medium-sized program comprising many
> hundreds of lines, broken up into several files and selectively imported,
> but do so in whichever language you know the *least*.  That's the way
> to learn, not though wars.  There's no way we can learn from one another
> by flaming.  And there's plenty to learn.  Everyone should do this.

As I have said before - if your only advice is "Go find out yourself, I
can't be bothered to tell you", then perhaps you should refrain from
posting.

Ian.



From greg.ewing at compaq.com  Tue Aug 17 17:36:54 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 18 Aug 1999 09:36:54 +1200
Subject: stat, chmod and RDONLY file (win32)?
References: <7pambk$3m8$1@ssauraaa-i-1.production.compuserve.com>
Message-ID: <37B9D5F6.2AC782E1@compaq.com>

"T. C. Mits" wrote:
> 
> mean that one must supply the c-run-time bit flags,
> documented elsewhere (where?), as mentioned in the open() API blurb, such as
> O_RDONLY.

No, O_RDONLY and friends are a completely different set
of modes (confusing, I know!)

The sort of mode you want here consists of 3 octal
digits, for owner, group and other. Each digit
contains read, write and execute permission bits
in that order.

So, to add write permission for the owner, for
instance, "or" the existing permissions with
octal 0200.

There's also a fourth octal digit to the
left which contains the setuid, setgid and
sticky bits, but these aren't relevant for what
you're doing here.

Hope that helps,
Greg



From patrick at c837917-a.potlnd1.or.home.com  Thu Aug 19 21:40:56 1999
From: patrick at c837917-a.potlnd1.or.home.com (Patrick Logan)
Date: Fri, 20 Aug 1999 01:40:56 GMT
Subject: 'if' as a _function_, in Scheme and Python
References: <7pi0kl$ko8$1@nnrp1.deja.com> <37bc9dd6.11041486@90.0.0.40> <Ze1v3.287$m84.4462@burlma1-snr2>
Message-ID: <Im2v3.7117$Hu6.1952@news.rdc1.wa.home.com>

In comp.lang.functional Barry Margolin <barmar at bbnplanet.com> wrote:

: Remember that then-branch and else-branch are thunks.  It doesn't
: matter whether you evaluate them, they don't do anything until you
: *call* them.  The above expression selects one of them and then
: calls it (notice the extra level of parentheses before "cdr").

: A more common implementation involves replacing #t and #f with
: functions:

: (define (if-true then-branch else-branch)
:   (then-branch))
: (define (if-false then-branch else-branch)
:   (else-branch))

BTW this is exactly the way "if" is implemented in Smalltalk.

  Boolean>>ifTrue: trueThunk ifFalse: falseThunk

...is a method that is redefined in Boolean's two subclasses, True and
False. For the class True, it is defined as...

  True>>ifTrue: trueThunk ifFalse: falseThunk

    "Return the result of the value of trueThunk"

    ^trueThunk value

and for class False...

  False>> ifTrue: trueThunk ifFalse: falseThunk

    "Return the result of the value of falseThunk"

    ^falseThunk value

Sending this message looks like...

    (a < b) ifTrue: [#yeah] ifFalse: [#boo]

...where (a < b) evaluates to either the singleton instance of True,
called true, or the singleton instance of False, called false.

The notation [<expression>] is a thunk like (lambda () <expression>)
in Scheme. With arguments a "lambda" in Smalltalk looks like...

    "This is a lambda that returns the sum of its arguments"
    [ :x :y | x + y]

-- 
Patrick Logan    patrickdlogan at home.com



From skip at mojam.com  Mon Aug  2 16:28:45 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon,  2 Aug 1999 15:28:45 -0500 (CDT)
Subject: hidden fields in HTML forms
In-Reply-To: <_Smp3.1$_j3.664@smithers.wenet.net>
References: <_Smp3.1$_j3.664@smithers.wenet.net>
Message-ID: <14245.64813.982756.757412@dolphin.mojam.com>

    Rick> Could anyone point me in the right direction of finding a cgi
    Rick> class that catches hidden fields sent to it from an HTML form?
    Rick> cgi.py doesn't recognize them;

Rick,

There's no reason for cgi.py to recognize anything special about hidden
fields.  They are only hidden from the web user, not the cgi script.  To
cgi.py, an input is an input.

If you want to fiddle around with the <INPUT> TYPE attribute and see what
the CGI script sees, set your <FORM>'s ACTION attribute to

    http://www.musi-cal.com/cgi-bin/query

That script uses cgi.FormContent() to grok form inputs and spits back
several bits of information about them and the environment in which the
script was executed.  If you give it a hidden input, e.g.:

    <form METHOD=GET ACTION="http://www.musi-cal.com/cgi-bin/query">
	<input TYPE=hidden name="hidden" value="You can't see it but I can!">
	<input TYPE=text name="visible" value="Users see this!">
	<input TYPE=submit>
    </form>

you'll see what I mean.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-475-3758



From philh at vision25.demon.co.uk  Thu Aug 26 16:38:12 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Thu, 26 Aug 99 20:38:12 GMT
Subject: Library philosophy
References: <37C57A80.D5C62704@prescod.net>
Message-ID: <935699892snz@vision25.demon.co.uk>

In article <37C57A80.D5C62704 at prescod.net>
           paul at prescod.net "Paul Prescod" writes:
> I am curious about the opinions of Python community members on the
> following issue:
> 
> I believe that Python's libraries are in many places too thin of a layer
> on top of the C library and are thus not very Pythonish.

The same thought has often occurred to me.

> Would it it make sense to start wrapping the low-level modules with
> high-level, easier to use modules that would be "the default" in Python
> 2?

This could be a useful idea, as long as the present libraries are
kept for compatibility.

I've written a class representing a Usenet article, which could form
the basis of a high-level nntp module.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From fdrake at acm.org  Tue Aug 24 15:19:54 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Tue, 24 Aug 1999 19:19:54 GMT
Subject: Python equivilent of pack()?
In-Reply-To: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com>
References: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com>
Message-ID: <14274.61530.251401.762453@weyr.cnri.reston.va.us>

Bit Nik writes:
 > My question is this: is there a functional equivilent to the Perl pack()
 > and unpack() functions in Python? Or even better, is there an easier way
 > to do what I want with Python? If not, how many people think this would be

  Take a look at the struct module; the pack() and unpack() functions
are exactly what you're looking for!
  (What names were you looking up in the index of the Library
Reference?  I'll add them to the index if they make sense to others.)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From arcege at shore.net  Fri Aug  6 10:25:05 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 06 Aug 1999 14:25:05 GMT
Subject: API question -- decref after adding to dictionary?
References: <rps64v$0$37nspbi$7gc@news.supernews.com> <379EBC8B.E36F020B@lemburg.com> <rpu849$1$37nspbi$7gc@news.supernews.com> <14239.11258.615882.396388@weyr.cnri.reston.va.us>
Message-ID: <5fCq3.701$EG4.94343@news.shore.net>

Fred L. Drake <fdrake at cnri.reston.va.us> wrote:

: Rich Salz writes:
:  > The API documentation doesn't seem to say what the functions do to
:  > object references.  Am I missing it, or do I really have to read the

: Rich,
:   This is a matter of completeness; the API docs just aren't "there"
: yet in this regard, and the time to devote to adding the information
: hasn't been available.
:   In the meanwhile, I'd certainly be glad to receive patches which
: added the information to the API reference.
:   I do hope that this isn't a deciding factor as to whether you use
: Python.  The discussions of reference counting in the API and
: Extending & Embedding manuals do try to explain the approaches taken
: in Python and reduce the pain significantly.  I've written a number of 
: C extensions myself, including one very early in my experience with
: Python, and haven't had a lot of difficulty with it.  (Note: I did
: *not* say I haven't had any difficulties!)

Fred, I haven't given up on my own extention writing section, just gotten
delayed researching the C++ side of it.  Should I just drop it?

  -Arcege




From greg.ewing at compaq.com  Wed Aug 18 19:00:01 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 19 Aug 1999 11:00:01 +1200
Subject: stat, chmod and RDONLY file (win32)?
References: <7pambk$3m8$1@ssauraaa-i-1.production.compuserve.com> <37B9D5F6.2AC782E1@compaq.com> <7pd13e$if1$1@ssauraab-i-1.production.compuserve.com>
Message-ID: <37BB3AF1.53A9D306@compaq.com>

"T. C. Mits" wrote:
> 
> Funny I associated the four octal flags with Unix, not win98 or
> NT.

The stat and chmod functions do their best to
emulate the unix ones on other platforms.
In the case of Wintel, probably not all the
mode bits are implemented, but at least the
ones that are should be in the right places.
Hopefully. With luck.

Greg



From pnpayne at swissonline.ch  Mon Aug 23 06:58:00 1999
From: pnpayne at swissonline.ch (Philip Payne)
Date: Mon, 23 Aug 1999 12:58:00 +0200
Subject: Python+Apache+Oracle+???Persistent database connection???
References: <37BDA33B.E8DFEBC5@swissonline.ch>
Message-ID: <37C12938.4A3EDAE7@swissonline.ch>

In case anyone is interested, the following solution was sent to me by
e-mail by Andrew Csillag:

Actually, there sorta is.  With PyApache 4.14 and up, there is something
that gets put into the builtin namespace, it's called __persistdict__,
which is a persistant dictionary in that it doesn't get wiped between
requests.  A point to note tho is that you really can only put in C
objects, cause python objects won't work (mostly because the module the
object came from was since destroyed).  I've done this with excellent
results with oracledb, with DCOracle it gets a bit trickier.  You wind
up having to do something like this (caveat programmer: I haven't
actually tried it, but I've worked with other objects that worked
similarly) to use the connection:

   to save the connection:
      #you might be able to get away without using __dict__ and perhaps
could
      #just use conn._d, but there's a setattr hook and I didn't really 
      #look at it to see what it did, so the below should work for sure
      __persistdict__['my_oracle_connection'] = conn.__dict__['_d']

   to use the saved connection:
      import DCOracle.ocidb
      class dummy: pass
      conn = dummy()
      conn._d = __persistdict__['my_oracle_connection'] 
      conn.__class__ = DCOracle.ocidb.ocidb

It it a bit ugly?  Yup, but it should work more or less.  While oracledb
isn't quite as nice in general, you can just stick it into the dict
(although, there is some mucking around with the dbi module to make IT
work).



From cgw at fnal.gov  Tue Aug 10 15:51:25 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 10 Aug 1999 19:51:25 GMT
Subject: What's a good way to get my IP from Python?
In-Reply-To: <37B07AB5.45556B52@cisco.com>
References: <37B04D5F.2F00EE6C@cisco.com>
	<14256.27817.224703.683095@buffalo.fnal.gov>
	<37B07AB5.45556B52@cisco.com>
Message-ID: <14256.33469.405885.755529@buffalo.fnal.gov>

A correction to the code I posted earlier:  change

    ifr = ifname+'\0'*(32-len(ifname))

to

    ifr = ifname+'\0'*(16-len(ifname))+chr(socket.AF_INET)+15*'\0'


As Aaron Rhodes pointed out to me, the version I originally posted
won't work on Linux 2.0 kernels.  It seems that the SIOCGIFADDR ioctl
wants to have the address family in the struct ifrequest on older
kernels, whereas kernel 2.2 doesn't care.  With the correction above,
the code I posted will work on either 2.0 or 2.2.  Also, there are no
differences in the struct ifreq, so no "offset tweaking" is necessary.




From jpl at digex.net.nospamplease  Thu Aug 19 15:35:02 1999
From: jpl at digex.net.nospamplease (John Landahl)
Date: Thu, 19 Aug 1999 12:35:02 -0700
Subject: No man pages? (was Re: More random python observations from a perl programmer)
In-Reply-To: <37bc4352@cs.colorado.edu>; from Tom Christiansen on Thu, Aug 19, 1999 at 11:48:02AM -0700
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu>
Message-ID: <19990819123502.A28466@digex.net>

On Thu, Aug 19, 1999 at 11:48:02AM -0700, Tom Christiansen wrote:
> In comp.lang.python, 
>     drew_csillag at geocities.com writes:
> :There are the python manuals which are excellent, available at:
> :http://www.python.org/doc/ (linked to from the homepage)
> :available in your choice of HTML, PostSript, PDF, Info, LaTeX and
> :windows help.
> 
> The fact that they did icky windows help and not man pages says
> something significant.

What's all this about no Python manpages?

  $ man python
  Reformatting page.  Wait... done



  Misc. Reference Manual Pages               PYTHON(10 April, 1998)



  NAME
       python - an interpreted, interactive,  object-oriented  pro-
       gramming language

Did this change relatively recently?  I'm running v1.5.2.

- John Landahl / jpl at digex.net



From bill.anderson at libc.org  Fri Aug  6 19:36:15 1999
From: bill.anderson at libc.org (Bill Anderson)
Date: Fri, 06 Aug 1999 17:36:15 -0600
Subject: Python and Red Hat Linux 6
References: <379CD547.22A10A50@postoffice.pacbell.net> <379CD5C2.5BA096A2@mail.danbbs.dk> <7nirk7$icq$1@oak.prod.itd.earthlink.net> <37A5F220.B83EEF99@boi.hp.com> <37A6420E.B0A24A24@mincom.com>
Message-ID: <37AB716F.726E530D@libc.org>

Martin Pool wrote:
> 
> Bill Anderson wrote:
> > Spud wrote:
> > >
> > > Yeah, using #!/bin/env/python should help, but Red Hat should support
> > > pound-bang (#!) anyway. You might want to check in with tech support.
> > >
> >
> > Uhmm ... it is not RedHat, but a shell thing, and if #! didn't work,
> > neither would #!/bin/env/python.
> 
> I think you mean
> 
>   #!/bin/env python

Bou are correct. Must've been a less-than-joyful day at work :-)



From rik-007 at dds.nl  Wed Aug  4 06:12:58 1999
From: rik-007 at dds.nl (Rik)
Date: Wed, 4 Aug 1999 12:12:58 +0200
Subject: Stupid question about .EXE
Message-ID: <7o93uk$sjrn$1@reader2.wxs.nl>

Hi everybody,

I've got a (stupid) question. I'm learning myself Python now, but I still
have a question:
How can you make .EXE files with python??? I'm using Python 1.5.2 (final)
for Windows.

Greets

Rik

P.S. Could somebody send me some source files of games made with python.
I've got BUS and PUB.





From paul at prescod.net  Sat Aug 28 06:22:38 1999
From: paul at prescod.net (Paul Prescod)
Date: Sat, 28 Aug 1999 06:22:38 -0400
Subject: Returning none
References: <37C55044.8CA02530@prescod.net>
		<37C5683D.91D61FD0@rubic.com>
		<14277.28442.44107.591412@dolphin.mojam.com>
		<37C70821.92E810FC@prescod.net> <14279.9024.401900.142461@dolphin.mojam.com>
Message-ID: <37C7B86E.647BA654@prescod.net>

Skip Montanaro wrote:
> 
> The same problem existed in the days when lint functionality was separate
> from the C compiler.  People just had lint targets in their Makefiles.  

Makefile? Isn't that what I'm using Python to get away from? Even in a
project with C components, I only run make when the C code changes, not
when the Python code changes.

 Paul Prescod



From m.faassen at vet.uu.nl  Mon Aug 30 10:02:36 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 30 Aug 1999 14:02:36 GMT
Subject: Upper limit?!
References: <37CA1E0F.FA2A05A3@megsinet.net> <014f01bef2bd$3fd09e90$f29b12c2@secret.pythonware.com> <37CA3EBD.4077AF18@megsinet.net>
Message-ID: <7qe2tt$p3k$1@newshost.accu.uu.nl>

David Oppenheimer <davidopp at megsinet.net> wrote:
[alpha allows longer integers]
> Wow, that's quite a difference! on an Alpha you can have a 19 digit number
> as opposed to a 10 digit number on my Pentium.

> So the limit isn't actually
> a factor of the programming language but a virtue (or limitation) of my
> processor.  I wonder if Intel knows about this?!

Intel knows for sure. :) Ever heard of 'merced'? Alpha is a 64 bits processor;
the intel 386, 486, and all pentiums are 32 bits processors. The merced will
be a 64 bits processor. Intel's a bit behind on converting to 64 bits
processors, but of course they have a huge legacy to deal with.

The maximum unsigned integer on a 32 bits processor (that can be held by
a single register and is the normal way to do integer calculations) is
2**32. For signed, one bit is needed to store the sign, so then it's 2**31.

For a 64 bit processor, this is 2**64 (2**63 for signed number). This is quite
a lot more. Really a *lot* lot more. Exponentials. :) 

This also really comes in handy when trying to address physical RAM. An 
address usually needs to fit in a physical register, and this means the
maximum amount of RAM a 32 bits chip can easily address is about 4 gigabytes.
(2**32 / (1024 * 1024)). This isn't a problem yet, but is slowly starting
to become one. With 64 bits we'll be happy for quite a while, it can accomodate
a couple of million *terabyte* just fine. About 16 thousand exabyte or so,
I think. :)

>   LOL  Makes me want to
> sell this hunk of crap Packard Hell P-75, get rid of Win98 and buy an Alpha
> running Linux.   That PSN number thing scares the hell outta me any ways,
> I'm either going AMD or Alpha next go around.

AMD's are 32 bit processors too, though, as far as I'm aware. Anyway, in
practice 32 bits are just fine for most purposes. If you need biiig numbers,
instead of integers you can use Python's long integers (which don't use
processor registers at all). If floating point numbers are fine you can
move on processor speed (if Python didn't slow things down :) -- Intel's
processors have a special floating point section for fast calculations
with these.

Regards,

Martijn




From kas at maps.magnetic-ink.dk  Sat Aug  7 01:13:18 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Sat, 7 Aug 1999 07:13:18 +0200
Subject: International characters and Python's string functions on Linux
References: <Pine.LNX.4.04.9908061223110.24989-100000@emerald.netskate.ru> <MPG.1215b842a85c7d2c989682@news.mop.no>
Message-ID: <slrn7qng3e.65g.kas@titan.magnetic-ink.dk>

Alexander Staubo wrote:
~~~~~~~~~
Beautiful name you've got there!  O:-)

>     from _locale import *
> ImportError: No module named _locale
> >>> 
> 
> This is Slackware Linux 2.2/libc6. Should I configure Python with some 
> kind of magic parameter for the locale stuff to work?

Go to the Modules directory of your python source tree, then edit the Setup
file to include the _locale module.


  //Klaus

-- 
???[ Magnetic Ink ]????????????????????????????????????????????????? ><> ???
???[ Key fingerprint: 2E37 EAED F848 815B 050A FBAA 8F68 DD66 B4B3 7EC4 ]???



From Friedrich.Dominicus at inka.de  Thu Aug 26 02:17:38 1999
From: Friedrich.Dominicus at inka.de (Friedrich Dominicus)
Date: Thu, 26 Aug 1999 08:17:38 +0200
Subject: C module leaking memory
References: <FH1wtI.C68@world.std.com>
Message-ID: <37C4DC02.32ADCC3A@inka.de>

Will Ware wrote:

The fun of manual memory management ;(

> 
> I'm working on a fairly complicated C module and it appears
> to be leaking memory. It runs a while, and then the interpreter
> gives up, saying 'MemoryError'. Are there any good pointers for
> debugging memory leaks? 

There are different packages or wrappers around malloc available. Try
e.g dmalloc and/or Electric Fence and ... 

One of the first points you can try is -Wall +W extra-options (for gcc )
and using lclint 

Dmalloc will keep book on all the allocation and freeing and this will
possibly help to find some leaks.

Regards
Friedrich



From xah at best.com  Sat Aug 21 20:09:44 1999
From: xah at best.com (Xah)
Date: Sat, 21 Aug 1999 17:09:44 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
	 <37b634a2@cs.colorado.edu>
Message-ID: <37bf3fc9$0$205@nntp1.ba.best.com>


----------
In article <37b634a2 at cs.colorado.edu>, Tom Christiansen 
<tchrist at mox.perl.com> wrote:


>      [courtesy cc of this posting mailed to cited author]
>
> In comp.lang.perl.misc, thehaas at my-deja.com writes:
> ...
> I think I can safely say, that the Perl
> :community is quite arrogant, especially to people new to it.
>
> It certainly didn't start out that way, and for many years Perl was
> famous for being precisely exactly the opposite of hostile.  It was
> about a programming community of mutual benefit.
>

> Think back to the 80s and early 90s.  The change, if it really happened,
> came with the upty-gazillion CGI script kiddies who couldn't program
> 2+2, yet who wanted *us* to write *them* these persistent, encrypted,
> crossplatform, multiscreen shopping carts for e-commerce, complete with
> dynamically-generated animated vanity counters, each and every time
> they stepped up to the feeder.

isn't that what perl republic wanted? If we want popularity, we cannot look
down on script kiddies. They are the ones who made Perl ubiquitous. Besides,
with Perl's DWIM (pronounced dim-wit) feature, much of what these kiddies
want are half done already.

> As a result, There's not much left of a spirit of a programming community
> of mutual benefit when it becomes numerically dominated by people who
> don't want to study or to work or EVEN TO PROGRAM, and who all seem to
> think you owe them something.

yeah I can grok that. Before I became a Perl expert, I was one of those
tedious perfectionist. Since then my brain has been freed. The Perl way(s)
has taught me the virtues of good programer. These days I can just fart and
food comes to my mouth, and I think everyone owes me something.

> When > 90% of the newsgroup postings are ill-formatted garbledy-gook
> that also fall into the category of not having bothered to have first
> checked the online docs or faqs that come with Perl, these things add up.
> And eventually, it risks breaking our spirit.  There's just no way to deal
> with the never-ending onslaught of non-programmer CGI script kiddies, who
> seem to outnumber the rest of us zillions to one.

Although I've been brain-washed for good, but one of my old trait --
documentation readings -- remained. I read all fags and your 4-camel books.
However, I no longer read them fastidiously. Perl has taught me to grok
docs, suck cheesiness, wallow in ambiguity, and swallow imprecision. Life
has never been as fun. Now I always think in contexts. As you can see, even
my writings are very context sensitive now.

> So what do you do?  You either bail out of this untenable situation, as
> Larry Wall has done, or else "customer-service battle fatigue" sets in.
> When that happens, people end up getting snapped at, sometimes mistaking
> an honest learner for yet another beggar looking for handouts.

Oh wait a min Tommy my leader. We are proud of calling ourselves mongers.
should the term 'beggar' not be used in a scary context? y'know, that might
muddy our reputation.

>  I dare
> say this would happen even in the best of families, given equivalent
> circumstances.

I concur to dare say that too. The big ass functional language family(ies)
never faltered on issues of quality, yet their collective newsgroup activity
is pitifully small. It's like a diamond lost in space.

> I do think that earnest individuals seriously wanting
> to work hard are still given the same helping hand they always got.

Yes. With you leading the way, we always benefits. It's like a prodding
cattle; there are always more than one way to get people going.

> Be careful what you wish for.
>
> --tom

Didn't _Larry_ said that sometimes in his life? Sorry I'm off topic here.

> --
> Emacs is a fine operating system, but I still prefer Unix. -me

Me 2. Don't you sometimes wish that you could just wipe the entire planet's
emac'ers off your face? Gnu is Not Unix my Ass, because it is.

Lastly, it is my wish that one day I'll soar into a Perl demigod status like
you. For one thing, you taught me how to advocate perl pretty well, have I?

with Perl virtues,

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 "The three principle virtues of Perl programers: mundaneness, sloppiness,
and fatuousness."




From robin at jessikat.demon.co.uk  Mon Aug  9 22:53:43 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Tue, 10 Aug 1999 03:53:43 +0100
Subject: Licenses...
In-Reply-To: <37AF12D3.71240BA7@lemburg.com>
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com>
 <37AA804D.9A1317C0@houbits.com> <934115482snz@vision25.demon.co.uk>
 <37ae686e.59511102@news.ricochet.net> <37AE9AC3.34007C4E@lemburg.com>
 <37AE9B42.4DF40758@lemburg.com> <SpY1dCA7Ctr3EwsH@jessikat.demon.co.uk>
 <37AF12D3.71240BA7@lemburg.com>
Message-ID: <B9rxoGA3Q5r3Ewuf@jessikat.demon.co.uk>

In message <37AF12D3.71240BA7 at lemburg.com>, M.-A. Lemburg
<mal at lemburg.com> writes
>Robin Becker wrote:
...
>
>If TCL were under GPL then you'd have to ship both TCL and
>your extension under GPL. But you're lucky: TCL is shipped under
>a more liberal license, so you can distribute your code in any
>form you like with or without TCL, as binary or source.
>
>Note that I just posted this because more and more GPLed extensions
>for Python are popping up -- if you want to include these in a
>product, you'll have to take some care to follow the extension's
>license. Otherwise, including the extension's code would be
>simply illegal. In most cases, the author will probably also
>distribute his/her code under a different (payed) license. In some
>cases (like the GNU readline module, for which Python includes
>an extension), the authors will probably not... e.g. shipping
>Python with linked in readline support (as DLL or statically)
>causes that Python interpreter to fall under GPL and with it,
>all extensions linked to it. The Python code run by the interpreter
>is not forced under GPL (otherwise all gcc compiled software
>would turn GPLed...).
...
my point was that since my extension can be used to link tcl to these
GPL'd dlls it might dynamically compromise the GPL status. Suppose that
python could be shipped with lazy links to readline would the potential
link matter?
-- 
Robin Becker



From gmcm at hypernet.com  Sun Aug  1 23:26:27 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 1 Aug 1999 22:26:27 -0500
Subject: Q: Why is __init__.py required
In-Reply-To: <7o2eeq$gs2@dfw-ixnews6.ix.netcom.com>
Message-ID: <1278561901-2477172@hypernet.com>

G. David Kuhlman writes:

> I have looked and looked for this.  Whenever I put an empty
> __init__.py in a package, I ask myself why I have to do this.  I
> read Guido's article on packages and it told me that I have to put
> an __init__.py in a directory in order for the directory to be
> recognized as a package.  But it doesn't say why.
> 
> I'm not asking for this behavior to be changed.  I like it the way
> it is.  I just want to know *why* it has to be that way.  In what
> way would the Python interpreter get confused if it didn't have
> __init__.py to distinguish package directories from non-package
> directories?

import a.b becomes:
 import a
 if a is a package:
    in this same directory, look for b
 else
    get this thing called b out of module a

> It bugs me when I don't know why.  My parents used to have to
> explain things to me like why I couldn't jump off the roof of the
> house.

they-sure-blew-that-one-<wink>-ly y'rs

- Gordon



From tchrist at mox.perl.com  Fri Aug 20 09:28:04 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 20 Aug 1999 06:28:04 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <012c01beeae7$f3582e20$f29b12c2@secret.pythonware.com> <37BD2BD8.5A8E@spider.com>
Message-ID: <37bd49d4@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Alex Maranda <amaranda at spider.com> writes:
:> >     So much for being typeless. Sigh.
:Tom, you might have done reasonably well the rest of your Python
:homework, but that is a very dumb statement. Tcl is typeless. Python is
:rather strongly typed and proud with that. 

I'm just reporting what I've read.  I wasn't aware that Mark Lutz was
considered a spreader of fictions.  I'd swear his book used that phrase
a lot.  I agree that dynamically typed is more sensible a term.

--tom
-- 
Perle, plesaunte to prynces paye/ To clanly clos in golde so clere,
Oute of oryent, I hardyly saye/ Ne proued I neuer her precios pere.



From gregm at iname.com  Tue Aug 31 21:37:12 1999
From: gregm at iname.com (Greg McFarlane)
Date: Wed, 1 Sep 1999 11:37:12 +1000
Subject: Grid (possible scrollable) in Tkinter?
In-Reply-To: <19990830214514.12034.qmail@www0j.netaddress.usa.net>; from Def P on 30 Aug 1999 at 09:45:14PM
References: <19990830214514.12034.qmail@www0j.netaddress.usa.net>
Message-ID: <19990901113712.54151@nms.otc.telstra.com.au>

On 30 Aug, Def P wrote:
> ... Tkinter does not have a
> scrollable listbox. To make one isn't that hard...

You are correct in saying that it is easy to write a simply scrollable
listbox (see below *).

To this you may like to add things like a horizontal scrollbar (as
well as the vertical); dynamic display (only display the scrollbars
when necessary); convenient methods to populate the listbox; an
optional label; a margin between the scrollbar and the listbox; single
and double click callbacks; easy and consistent configurability of all
component options, etc.  You may then want to use all this as part of
a dropdown listbox (combobox).  If you want these features, it may
save you some pain to reuse an existing library, such as Pmw.

> ...Is building these things worthwile or shouldn't I bother
> and look into libraries like Pmw? For me, the ideal situation would be if I
> could distribute my programs without requiring external libraries.

If you are worried about having a dependency on Pmw, it is easy to
concatenate all of the Pmw modules you require into a single file and
include this with your software distribution.  A script to do this can
be found in bin/bundlepmw.py in the Pmw distribution at
<http://www.dscpl.com.au/pmw/>.


* Simple example of scrollable listbox:

    l = Tkinter.Listbox()
    l.pack(side = 'left', fill = 'both', expand = 1)
    sb = Tkinter.Scrollbar(command = l.yview)
    sb.pack(side = 'left', fill = 'y')
    l.configure(yscrollcommand = sb.set)

-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
                   of precipitation in some areas.



From moshez at server.python.net  Thu Aug 19 01:19:47 1999
From: moshez at server.python.net (Moshe Zadka)
Date: Thu, 19 Aug 1999 01:19:47 -0400 (EDT)
Subject: PIGgies?
In-Reply-To: <slrn7rn3ej.ugg.kaz@ashi.FootPrints.net>
Message-ID: <Pine.LNX.4.10.9908190118060.28527-100000@starship.python.net>

On Thu, 19 Aug 1999, Kaz Kylheku wrote:

> On Wed, 18 Aug 1999 17:15:23 -0700, John Landahl <jpl at digex.net.nospamplease>
> wrote:
> >These are my opinions, of course, and I graciously solicit the opinions of
> >others.  What do you think of the term "PIGgies"?  Do you feel that there is
> 
> Little piglets are cute animals. Besides, you obviously haven't seen an angry
> boar charging at you at 100 miles per hour.

<philosophy>
If a boar and a penguing would charge at each other,
would anyone hear?
</philosophy>

well-now-we-have-a-real-python-mascot-ly y'rs, Z.




From tchrist at mox.perl.com  Fri Aug 20 09:07:09 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 20 Aug 1999 06:07:09 -0700
Subject: More random python observations from a perl programmer
References: <37BC6986.A66B044E@starmedia.net> <199908191634.KAA01333@jhereg.perl.com> <87d7wiaf8q.fsf@deneb.cygnus.qad.org>
Message-ID: <37bd44ed@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Florian Weimer <fw at s.netic.de> writes:
:Manpages are most useful for languages which do not support namespaces
:(so you can hit `K' or `M-x man RET RET', and you get the right page).
:Perl fits into this more than Python (which heavily uses methods on
:builtin objects).  Unfortunately, `man exists' (or `perldoc exists')
:doesn't work with my Perl installation either, 

I really hate that.  You're right about perlfunc being insanely long.
I've always argued against it, and there's a script in the distribution
kid that *will* create things like /usr/local/perl/man/man3/exists.3,
but I haven't managed to get my lobbying power worked up enough to push
through to make that the default.

--tom
-- 
	 "If you don't know where you are going, any road will take you there."
	 Lewis Carroll, "Alice in Wonderland".



From bwarsaw at cnri.reston.va.us  Thu Aug 19 15:13:42 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 15:13:42 -0400 (EDT)
Subject: More random python observations from a perl programmer 
References: <14268.15201.81468.481452@bitdiddle.cnri.reston.va.us>
	<199908191713.LAA02152@jhereg.perl.com>
Message-ID: <14268.22374.331241.721504@anthem.cnri.reston.va.us>

>>>>> "TC" == Tom Christiansen <tchrist at jhereg.perl.com> writes:

    TC> I only have the Lutz book.  I don't have manpages. :-(

That's okay, I only have the first edition Programming Perl book :)

-Barry



From jbkerr at sr.hp.com  Mon Aug  2 16:52:37 1999
From: jbkerr at sr.hp.com (Jim Kerr)
Date: Mon, 02 Aug 1999 13:52:37 -0700
Subject: stuck on COM interface to Word
Message-ID: <37A60515.370B5085@sr.hp.com>

  I'm trying to write a python script to extract some properties from a set
of Word documents. The attributes I'm interested in are things like author's
name, last print date, keywords, etc.

  This is what I have so far:

from win32com.client import constants,Dispatch
import string,os

app = Dispatch('Word.Application.8')
app.Documents.Open(r"C:\My Documents\myMemo.doc")
doc = app.ActiveDocument
print "Last printed date: ",  # ?? what goes here ??

  I tried using combrowse.py to figure out what the properties of the doc are,
but I couldn't find anything that looked relevant.

  Is there any web-based info on COM interfaces to various programs? Or do you
have to buy Visual Basic books and translate for yourself?

  Thanks for the help!

						Jim
-- 
Jim Kerr
Hewlett-Packard
Santa Rosa Systems Division
1400 Fountaingrove Pkwy, MS 3USZ
Santa Rosa, CA 95403



From noreply at domain.com  Tue Aug  3 13:07:40 1999
From: noreply at domain.com (Brian)
Date: Tue, 3 Aug 1999 10:07:40 -0700
Subject: Python in web page
References: <37a70f28.96313821@midpoint>
Message-ID: <xoFp3.28$o3.7003@news.premier1.net>

Hi Tim,
You could use JPython to generate Java applets.  I have a recent posting
myself requesting feedback on JPython, but no responses yet.  I just started
looking at it, but so far it looks pretty good.  Take a look:
http://www.jpython.org/
HTH,
Brian

Timothy Dixon wrote in message <37a70f28.96313821 at midpoint>...
>Is it possible to host a Python application in a web page?  TCL has a
>plug-in for IE and Netscape; is there something similar for Python?
>





From dfan at harmonixmusic.com  Thu Aug 19 10:04:46 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 19 Aug 1999 10:04:46 -0400
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <y0jiu6edoa9.fsf@vier.idi.ntnu.no> <xj0d7wl9acd.fsf@cyclone.harvard.edu> <37BAB643.9D4E861E@strs.co.uk>
Message-ID: <wkyaf7kgsh.fsf@turangalila.harmonixmusic.com>

Ian Clarke <I.Clarke at strs.co.uk> writes:

| I did some timing tests using the code I have attached below, here are
| the results (in seconds) for sorting a list of 200 items 10 times:
| 
| Dan Schmidt's solution (using sort): 14.98
|
| def s1(list):
|     """Dan Schmidt's solution - (mis)using sort"""
|     def randomize(a,b):
|         return random.choice([-1, 0, 1])
|     list.sort(randomize)
|     return 

That was Jeremy Hylton's solution.  I just complained about it!

it's-libel-I-tell-you-ly y'rs,
Dan

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From fredrik at pythonware.com  Fri Aug 20 08:56:42 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 12:56:42 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <37bc313b@cs.colorado.edu> <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us> <019301beeaea$004c4d80$f29b12c2@secret.pythonware.com> <199908201220.GAA24045@jhereg.perl.com>
Message-ID: <009501beeb0b$740b0740$f29b12c2@secret.pythonware.com>

Tom Christiansen <tchrist at mox.perl.com> wrote:
> Don't you understand your gain?  How dare you punch a gifthorse in the
> mouth--repeatedly?  If these petty slams and disdainful arrogances are the
> kind of response typical of your community, this bodes not well for you.

as one of the more prolific posters on this list/group, and
one who reads lots of other python lists, and speaks to many
python newbies both via mail and in person, I can assure you
that you are not behaving like the typical Python newcomer
in any fashion whatsoever.  it's okay to be different, but the
"I'm better, and you folks are not really worthy" stuff won't
get you nowhere.  face it: you're not, and we are.  calming
down a little may help you realise that.

on the other hand, since I don't know you in person, and
since the followups I quoted in an earlier post (as well as
the mails I recently found in my mailbox) didn't match the
positive tone and the content of your original post, I can
only hope that it's someone else who has hijacked your
computer.

> The more the pissier, you know.  I'll try to keep that in mind as I continue
> to attempt to sift through this dreck to find the gems.

good luck!

</F>




From garry at research.att.com  Mon Aug 30 12:07:39 1999
From: garry at research.att.com (Garrett G. Hodgson)
Date: Mon, 30 Aug 1999 16:07:39 GMT
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com> <37c748ad.1418655478@news.triode.net.au>
Message-ID: <37CAAC4B.9F1BCCFC@research.att.com>

John (Max) Skaller wrote:

>         Python is better (IMHO) at a larger scale of problem than
> some other scripting language because:
...
>         b) it is built on a few simple concepts, so it is
> relatively easy to _reason_ about parts of a program
> without having to hold all the code in your brain at once.

this is perhaps the clearest description i have yet seen
that explains why i stopped using perl and switched to
python.


-- 
Garry Hodgson			"Hey, mister, can ya tell me,
garry at sage.att.com		where a man might find a bed?"
Software Innovation Services	He just grinned and shook my hand,
AT&T Labs			"No", was all he said.



From bennettt at am.appstate.edu  Mon Aug 23 10:14:04 1999
From: bennettt at am.appstate.edu (TMGB)
Date: Mon, 23 Aug 1999 10:14:04 -0400
Subject: NEWBIE:Let's try that again..Help w/path
References: <7pm0o4$e0h$1@nnrp1.deja.com>
Message-ID: <37C1572C.A13C9321@am.appstate.edu>

Not knowing your experience, are you familiar with the DOS autoexec.bat
if so this should work for you.  If not you should be able to find
autoexec.bat in the c:\ directory, it is an ascii or ansi text file
depending if it has been edited before or not.


This is what was put in my autoexec.bat file when installing python and
python works on my PC.

set PYROOT=c:\py15
PATH %PYROOT%;%PATH%
set
PYTHONPATH=.;%PYROOT%;%PYROOT%\lib;%PYROOT%\lib\plat-win;%PYROOT%\lib\lib-tk;%PYROOT%\PIL;%PYROOT%ext\pyds;%PYROOT%ext\lib;
set TCL_LIBRARY=%PYROOT%\tcl8.0\library
set TK_LIBRARY=%PYROOT%\tk8.0\library



You will need to change PYROOT to 
set PYROOT=c:\progra~1\python
if that is where you installed python

and if you installed tcl8.0 and tk8.0 to a different directory than the
default also, you will need to change the path for those.  I think this
should work for your case.  My installation is Python build 125

Thomas

chinaski_py at my-deja.com wrote:
> 
> I'm new and need help with my Python path.When I go to open a Python
> file (.py) it won't open .I go to (on Win98):
> start
> run
> sysedit
> and here is my path:
> path %path%;C:\Perl\bin;path=C:\PROGRA~1\Python\;%path%
>                 Any help on this would be appreciated,
>                                 G.Norton (chinaski_py)
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

-- 
----------------------------------------------------------------------
Thomas McMillan Grant Bennett           Appalachian State University
Computer Consultant II                  University Library
bennettt at am.appstate.edu               
http://www.library.appstate.edu/admin/
Voice:  828 262 6587			FAX:    828 262 3001

Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
operating system that was originally coded for a 4-bit microprocessor.
 - Chris Dunphy     Boot Magazine



From thomas.heller at ion-tof.com  Thu Aug  5 13:51:23 1999
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 5 Aug 1999 19:51:23 +0200
Subject: Gordons installer or imputil problem?
Message-ID: <002101bedf6b$21c64810$4500a8c0@thomasnotebook>

Gordon,
I tried out your installer (the new beta) and am really impressed.
IMHO the best part is parsing the binary files (dlls, pyds) for dependecies.

I noticed a strange problem however, and I do not know where to look
for the solution.

Consider a simple python package:

directory 'package':
    file '__init__.py'
    file 'mod.py'

and the following scripts:

-----<script a>-----
print "Importing package"
import package
print "Importing package.mod"
import package.mod
print "Importing mod from package"
from package import mod
-----<EOF script a>-----

-----<script b>-----
print "Importing package"
import package
print "Importing mod from package"
from package import mod
print "Importing package.mod"
import package.mod
-----<EOF script b>-----

Since 'package' as well as 'package.mod' is imported
and modulefinder.py has no problems finding the needed files
and a.exe and b.exe are generated correctly.

Running the scripts shows the problem:

  C:\>a
  Importing package
  Importing package.mod1
  Importing mod1 from package

  C:\>b
  Importing package
  Importing mod1 from package
  Traceback (innermost last):
    File "C:\launch.py", line 5, in ?
      from package import mod1
  ImportError: cannot import name mod1

  C:\>

The only difference between a.py and b.py is that in the first case
 'import package.mod' is executed before 'from package import mod',
while in the other case the order is different.
Since I'm working with packages a lot, and normally use 'from package import
module'
I would stronly prefer to find a solution other than rewriting all the
import statements.


Thomas Heller
ION-TOF GmbH




From hyoon at bigfoot.com  Tue Aug 17 13:42:11 1999
From: hyoon at bigfoot.com (Hoon Yoon)
Date: Tue, 17 Aug 1999 13:42:11 -0400
Subject: Bug in struct module?
References: <000001bee869$8e9cbf80$6e2d2399@tim>
Message-ID: <37B99EF3.6376E1DB@bigfoot.com>

Everyone,

  Thanks that solves it. I am just beginning to dive into the binary
files. And platform was NT, I will remember to put more details in the
future (sorry).

Hoon,

-- 
*****************************************************************************
S. Hoon Yoon (Quant) Merrill Lynch Equity Trading, 
yelled at yahoo.com hoon at bigfoot.com(w)
"Miracle is always only few standard deviations away, but so is
catastrophe."
* Expressed opinions are often my own, but NOT my employer's.
"I feel like a fugitive from the law of averages."    Mauldin
*****************************************************************************
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vcard.vcf
Type: text/x-vcard
Size: 202 bytes
Desc: Card for Hoon Yoon
URL: <http://mail.python.org/pipermail/python-list/attachments/19990817/a680222a/attachment.vcf>

From paul at prescod.net  Fri Aug 27 17:50:25 1999
From: paul at prescod.net (Paul Prescod)
Date: Fri, 27 Aug 1999 21:50:25 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net>
		<37C5683D.91D61FD0@rubic.com> <14277.28442.44107.591412@dolphin.mojam.com>
Message-ID: <37C70821.92E810FC@prescod.net>

Skip Montanaro wrote:
> 
> 
> How about a compromise?  Perhaps Aaron Watters' PyLint can be modified to
> bitch about functions without a return as the last statement?

I would accept this type of compromise if Python had a way of
integrating "lint" into the edit/compile/run cycle. Most developers
don't remember to "pylint" six modules just because they changed each of
them in the last hour.

A lint "hook" might be enough to allow the Python community to get
started implementing cool warning types. But I haven't thought that
true...

 Pau Prescod



From robin at jessikat.demon.co.uk  Fri Aug 27 14:10:38 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Fri, 27 Aug 1999 19:10:38 +0100
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu>
 <37c126a6$0$813@newsreader.alink.net>
 <slrn7s2lb3.k6b.abigail@alexandra.delanet.com>
 <7putsq$3ql$1@lycaeus.calstatela.edu>
 <MPG.122fcf84a811c6099896e8@news.melbpc.org.au>
Message-ID: <hAxJjJAeStx3Ew5V@jessikat.demon.co.uk>

In article <MPG.122fcf84a811c6099896e8 at news.melbpc.org.au>, Factory
<faqtori at hotmail.com> writes
>In article <7putsq$3ql$1 at lycaeus.calstatela.edu>, ramune at bigfoot.com 
>says...
>> In article <slrn7s2lb3.k6b.abigail at alexandra.delanet.com>,
>> Abigail <abigail at delanet.com> wrote:
>> [snip amusing rant]
>> >Yeah, those math texts from the 16th and 17th century, they are sooooo easy
>> >to read, because they lack all the symbols. Not to mention the original
>> >Euclid and those other Greeks, who had no symbols at all!
>> 
>> <GRIN>
>> Oh, but they're written with only symbols!  All those confusing greek symbols
>> that mathematicians use are the basic building blocks for the language!  Why
>> didn't they just use English?
>> </GRIN>
>
>  Hmm that might be a bit off, the anchient Greeks had no concept of 
>variables, thus you would never see f(x)= 2y in any old greek 
>mathematics. IIRC variables were discovered(?) by Al-Jabar sometime 
>after.
>  Thus the Greeks certainly did not use symbols, and it held back 
>mathematics. Not that this applies to python vs. perl.
C:\Python\Installs>python shaney.py \tmp\junk.txt
Anyway, I'll post this with my comments as I originally started to enter
them.
Great work.
In geek religion wars, there are two types of people.
One of them as a way to increase your understanding.
(Of course, while the abstract material in CS can come in handy, if
they're taught wrong, what's the point?) There's quite a bit slow on the
uptake.
I should have read all the way through.
And I looked at your web site.
I must admit you write great parody.
Anyway, I'll post this with my comments as I originally started to enter
them.
Great work.

-- 
Robin Becker



From webmaster at python.org  Fri Aug 20 13:35:59 1999
From: webmaster at python.org (webmaster at python.org)
Date: Fri, 20 Aug 1999 13:35:59 -0400 (EDT)
Subject: Updated Python conference information
Message-ID: <14269.37375.119032.727000@anthem.cnri.reston.va.us>

We finally have information about the 8th International Python
Conference up on the Web.  This included dates, hotel, and fees.
Please see

    http://www.python.org/workshops/2000-01/index.html

for more information.  Details about some of the other sessions
(demos, posters, developers' day) will be added soon.

IMPORTANT: Note that the deadline for paper and tutorial submissions
is just a little over a month away.  See the Call for Papers for
details.  The deadline is September 30.

-Barry



From mpersico at erols.com  Sun Aug 15 15:22:02 1999
From: mpersico at erols.com (Matthew O. Persico)
Date: Sun, 15 Aug 1999 15:22:02 -0400
Subject: Wacky Programming Tales (Was Re: Why use Perl when we've got 
 Python?!)
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <slrn7rdgo2.a5.abigail@alexandra.delanet.com>
Message-ID: <37B7135A.3696C630@erols.com>

I once had to take over a MAJOR project from a consultant who was leaving.
This happened one week before my honeymoon. How major? Well it wrote and
sent settlement information for about 10-20% of the daily volume of equity
transactions in the entire United States.

Anyway, it took about six months of re-programming to evolve this morASS
(emphaisis mine) of code into something that behaved in a deterministic
fashion. As I hacked way at the underbrush, I kept coming upon all sorts of
references to a function called ocrap().  Eventually, I found the source
for the function (instead of just linking in the existing library) and
discovered it went something like this:

int ocrap(char *s) {

	printf("Oh crap, what am I doing here: %s\n",s);
	exit (5);
}


Abigail wrote:
> 
> Brad Howes (bradh at mediaone.net) wrote on MMCLXXV September MCMXCIII in
> <URL:news:wjy7lmxbwh7.fsf_-_ at bradh.ne.mediaone.net>:
> &&
> && Hey! I actually worked on a project where someone did this! Also in their
> && funky include file was
> &&
> &&   #define AND &&
> &&   #define OR  ||
> &&
> && It was the most bizarre thing I ever saw -- not so much the include file,
> && but the resulting code. As if a Pascal moths had come in, eaten at some
> && code and left little Wirth fragments sprinkled around. Has anyone else
> && encountered such strange programming behavior?
> 
> Well, uhm, yeah. I once wrote a C program implementing red-black trees
> that almost read like English. Except for parens and the one #include
> line, no punctuation chars where needed. It had things like:
> 
>       if (I am red and my uncle is black)
>       we do   ....
> 
> Unfortunally, that code didn't survive an extreme quota regime.
> 
> Abigail
> --
> perl -MNet::Dict -we '(Net::Dict -> new (server => "dict.org")
>                        -> define ("foldoc", "perl")) [0] -> print'
> 
>   -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
>    http://www.newsfeeds.com       The Largest Usenet Servers in the World!
> ------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----

-- 
Matthew O. Persico
    
You'll have to pry my Emacs from my cold dead oversized
   control-pressing left pinky finger. -- Randal L. Schwartz



From mal at lemburg.com  Sat Aug 21 12:26:57 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sat, 21 Aug 1999 18:26:57 +0200
Subject: [ANNOUNCE] M2Crypto 0.01
References: <37bd6fa4.0@news.cyberway.com.sg>
Message-ID: <37BED351.55E7DA98@lemburg.com>

Ng Pheng Siong wrote:
> 
> Hello,
> 
> I am pleased to announce the release of M2Crypto 0.01;
> the package is available at http://www.post1.com/home/ngps
> 
> Following is the README:
> 
>  The M2 Python Crypto Toolkit 0.01
> -----------------------------------
...

Could you please tell me where that server is located and whether
it is legal to download the software from Germany ? If it is
US based then we have the same problem as I had with the other
two OpenSSL interfaces: it's not legal to export them from the US
since even code linking at crypto software (in law speak: providing
crypto specific hooks) is understood as being contrary to the
US export regulations (at least that's what the Lotus Notes
chief engineer said about it and he should know).

Anyway, mxCrypto development will continue to eventually include
most APIs of OpenSSL.

BTW, a way to speed up this work is by funding it: OpenSSL is really
hard to wrap right because of its poor documentation and the task is
rather time consuming.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   132 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From ron at library.ucsf.edu  Thu Aug 19 16:47:01 1999
From: ron at library.ucsf.edu (Ron Klatchko)
Date: Thu, 19 Aug 1999 13:47:01 -0700
Subject: PIGgies?
References: <19990818171523.A22207@digex.net>
Message-ID: <37BC6D45.BE5ED382@library.ucsf.edu>

John Landahl wrote:
> Am I the only Python enthusiast that has a problem with
> the term "PIGgies"?

If I can't dance, I don't want to be part of your revolution. -Emma
Goldman.

moo
----------------------------------------------------------------------
          Ron Klatchko - Manager, Advanced Technology Group           
           UCSF Library and Center for Knowledge Management           
                        ron at library.ucsf.edu



From da at ski.org  Fri Aug  6 12:36:18 1999
From: da at ski.org (David Ascher)
Date: Fri, 6 Aug 1999 09:36:18 -0700 (Pacific Daylight Time)
Subject: [Image-SIG] Proper application of the buffer interface
In-Reply-To: <37AAA7FE.FFE0FD2@lemburg.com>
Message-ID: <Pine.WNT.4.04.9908060933080.213-100000@rigoletto.ski.org>

On Fri, 6 Aug 1999, M.-A. Lemburg wrote:

> Fred L. Drake, Jr. wrote:
> > 
> > David Ascher writes:
> >  > Would it also make sense to make struct.pack() return a buffer object
> >  > instead of a string?
> > 
> >   Sounds like it to me.  If Guido agrees, I can make the changes.
> 
> Wait... what's so bad about buffer(struct.pack()) ? Strings already
> know the buffer interface, so this works just fine already.

It's not terrible, but it seems suboptimal.  It strikes me that
struct.pack() explicitely shouldn't create strings, it should create
arrays of bytes.  Let's ask the question the other way -- what's wrong
with having struct.pack() return buffers instead of strings?  The only
thing I can think of is that type(struct.pack()) is no longer StringType.
Maybe that's enough to push the change to 2.0, I'm not sure either way.

--david






From pduffin at mailserver.hursley.ibm.com  Mon Aug  9 05:09:57 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Mon, 09 Aug 1999 10:09:57 +0100
Subject: Linux why not fully dynamic?
References: <1278177653-25589496@hypernet.com>
	 <37AAF3D5.141B55EB@bioreason.com> <37AB003C.28D9@mailserver.hursley.ibm.com> <Yvk34MAiqCr3Ewcx@jessikat.demon.co.uk>
Message-ID: <37AE9AE5.48EE@mailserver.hursley.ibm.com>

Robin Becker wrote:
> 
> In article <37AB003C.28D9 at mailserver.hursley.ibm.com>, Paul Duffin
> <pduffin at mailserver.hursley.ibm.com> writes
> > [snip]
> Paul was a major force behind the stubs mechanism in Tcl (I believe
> because of failings in AIX). The stubs mechanism allows versions to be
> more independent and overcomes deficiencies in inter extension linking.
> There is certainly less of this in python as extensions tend to
> implement specific python objects and normally don't refer across
> directly to the internals of other extensions.
> 

I cannot remember exactly why I did it now but I seem to remember
something about Jean-Claude Wippler holding a swirling disk in front
of my face ..... at -|

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From mwh21 at cam.ac.uk  Tue Aug 24 18:07:51 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 24 Aug 1999 23:07:51 +0100
Subject: Converting from perl
References: <7pujub$75v$1@nnrp1.deja.com>
Message-ID: <m3ogfwrfwo.fsf@atrus.jesus.cam.ac.uk>

sp00fD <sp00fD at yahoo.com> writes:

> I'm converting to python from perl and just created my first python
> program, it's actually a rewrite of a perl program that I had.  It's
> kind of stupid, but I find rewriting things in other languages is one of
> the best ways to learn.  What I'd like is for someone to tell me what's
> screwed up and what I need to work on.  The majority of the script is
> held in a class, but being a newbie to the world of OOP also, I'm not
> sure if the way I've handled things is optimal, so _any_ suggestions
> would be appreciated, as well as any "Why the hell did you create a
> class that way"'s or "You're all screwed up, you need to change
> this.."'s.  Anyway, here it is...

Looks OK to me. A couple of small points...

>         def get_hosts(self):
>                 if self.args() == "AIX":
>                         self.hosts = ("athena", "dev", "detroit", "la",
> \
>                                       "laredo", "miami", "newark",
> "sat", \
>                                       "savannah", "seattle", "titan")
>                 elif self.args() == "SUN":
>                         self.hosts = ("eclipse", "hercules", "scdev", \
>                                       "scprod", "sol")
>                 elif self.args() == "BACKUP":
>                         self.hosts = ("eclipse", "hercules", "scdev", \
>                                       "scprod", "sol", "athena")
>                 else:
>                         self.hosts = self.ask_for_hosts()

It might be easier to do this using a dictionary:

arg2hosts = {
    "AIX":("athena","dev","detroit","la","laredo","miami",
           "newark","sat","savannah","seattle","titan"),
    "SUN":("eclipse", "hercules", "scdev","scprod","sol"),
    "BACKUP":("eclipse", "hercules", "scdev","scprod", "sol", "athena")
}

try:
    self.hosts = arg2hosts[self.args()]
except KeyError:
    self.hosts = self.ask_for_hosts()

Your code could execute self.args() reoeatedly which is unecessary.

> 
>         def copy(self):
>                 _stuff = ()
>                 _stuff = self.files()
> 
>                 try:
>                         for host in self.get_hosts():
>                                 print "Sending %s to %s %s" % \
>                                 (_stuff[0], host, _stuff[1])
> 
>                                 os.system("rcp %s %s:%s" % \
>                                 (_stuff[0], host, _stuff[1]))
>                 except os.error:
>                         print "Error during system() function - oops"

Note that this will not catch errors in executing the command passed
to os.system - you need to look at the return code for that.
 
> c = CopyClass()
> 
> c.copy()

To be honest, I don't see a lot of value in wrapping this up in a
class. For a straight through shell scripty type job like this I
generally don't bother. Python is not about ramming methodologies down
your throat.

> </PRE>
> 
> Hope that PRE works

Hmm, seemed unecessary here...

HTH,
Michael



From gmcm at hypernet.com  Mon Aug 30 09:37:47 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 30 Aug 1999 08:37:47 -0500
Subject: fo.flush() & fo.write() on windows
In-Reply-To: <000001bef280$597f4040$a22d153f@tim>
References: <37C9C25F.DE0BEEAB@rl.gov>
Message-ID: <1276106069-51511751@hypernet.com>

[The Timbot]
> [Roberts, Robert J]

> >     foSize = os.stat(foName)[6]     # if foSize = 100...
> >     fo.write('Some string.')             # fo.write() returns None
> >     fo.flush()
> >     foSize = os.stat(foName)[6]    # ...foSize STILL = 100.
> 
> Ah -- *that's* Windows <wink>!

Curiously enough, NT4 (using NTFS) behaves much more sensibly:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam 
>>> f = open('temp.txt','a') 
>>> f.write('some string') 
>>> import os 
>>> os.path.getsize('temp.txt') 
0 
>>> f.flush() 
>>> os.path.getsize('temp.txt') 
11 
>>>

To the original poster: using f.tell() (after f.seek, if necessary) 
is a much more sensible (but not gotcha-free*) way of dealing with 
the size of an open file(**).

(*) It's not gotcha-free in text mode because it counts line endings 
as one (logical) char, while in fact you get 2 physical chars. 

(**)On really old Windows file systems, text files grow (from the
os.path perspective) in chunks. 

- Gordon



From pandeng at telepath.com  Fri Aug 20 02:01:45 1999
From: pandeng at telepath.com (Steve Schafer)
Date: Fri, 20 Aug 1999 06:01:45 GMT
Subject: 'if' as a _function_, in Scheme and Python
References: <7pi0kl$ko8$1@nnrp1.deja.com> <37bc9dd6.11041486@90.0.0.40> <Ze1v3.287$m84.4462@burlma1-snr2>
Message-ID: <37beeec3.31761662@90.0.0.40>

On Fri, 20 Aug 1999 00:24:25 GMT, Barry Margolin
<barmar at bbnplanet.com> wrote:

>No.  Remember that then-branch and else-branch are thunks.  It doesn't
>matter whether you evaluate them, they don't do anything until you *call*
>them.

Yes, but that doesn't negate what I said--it's not equivalent to IF
(since IF doesn't require that you wrap the expressions in thunks). 

Essentially, you're changing the problem to fit your solution. I guess
I don't really see any point in doing that.

-Steve




From see-sig at wavedivision.com  Mon Aug  2 17:54:48 1999
From: see-sig at wavedivision.com (Rick Robino)
Date: Mon, 02 Aug 1999 21:54:48 GMT
Subject: hidden fields in HTML forms
References: <_Smp3.1$_j3.664@smithers.wenet.net> <14245.64813.982756.757412@dolphin.mojam.com>
Message-ID: <Isop3.2$_j3.673@smithers.wenet.net>

Skip Montanaro <skip at mojam.com> wrote:

Thanks Skip, I appreciate the quick response (I would actually like
to use this thing myself!).

>     Rick> Could anyone point me in the right direction of finding a cgi
>     Rick> class that catches hidden fields sent to it from an HTML form?
>     Rick> cgi.py doesn't recognize them;

> Rick,

> There's no reason for cgi.py to recognize anything special about hidden
> fields.  They are only hidden from the web user, not the cgi script.  To
> cgi.py, an input is an input.

Understood - everything is there, but cgi.FieldStorage seems to only
accept keys from the text types of the form input.

> If you want to fiddle around with the <INPUT> TYPE attribute and see what
> the CGI script sees, set your <FORM>'s ACTION attribute to

>     http://www.musi-cal.com/cgi-bin/query

I've been using the apache cgi script logging facility and can see 
the whole string, including hidden fields, being passed. But thanks
for the tip here, I'll check it out.

> That script uses cgi.FormContent() to grok form inputs and spits back
> several bits of information about them and the environment in which the
> script was executed.  If you give it a hidden input, e.g.:

>     <form METHOD=GET ACTION="http://www.musi-cal.com/cgi-bin/query">
> 	<input TYPE=hidden name="hidden" value="You can't see it but I can!">
> 	<input TYPE=text name="visible" value="Users see this!">
> 	<input TYPE=submit>
>     </form>

> you'll see what I mean.

Hmmm... I did use your test - "query" and saw the hidden field output:

    Raw Input:
      hidden=You+can%27t+see+it+but+I+can%21&visible=Users+see+this%21

    Form Fields:
      hidden: ["You can't see it but I can!"]
      visible: ['Users see this!']

Maybe my problem is that the main class my data is coming from doesn't
ever use the raw_input method because I'm only using cgi.FieldStorage().
and not FormContent(). What I've done (remember, I'm just starting
off), was to rip off the FormData class used as an example in the
"Learning Python" book. Here it is, modified to ignore blank fields:

class FormData:
	""" A repository for information gleaned from a CGI form """
	def __init__(self, formdict):
		for fieldname in self.fieldnames:
			if not form.has_key(fieldname):
				bad_web()
			else:
				setattr(self, fieldname, form[fieldname].value)

(thanks Mark Lutz and David Ascher!)

When I use this, in classes I haved defined for emailing and formatting
the data as HTML output, it chokes because the form.has.key(fieldname)
test does not pass. Thats my problem: the whole string gets sent to the
cgi, but I'm missing something from my base class definition that pays
attention to field types other than "Text". If you want, I can send
the cgi and the form (149 and 161 lines, respectively).

This is probably where I need to learn more, and the FormContent class
is probably the class that will help me here. However, the docstring for
that class says:

    """This class is present for backwards compatibility only."""

What do you think? Should I use FormContent or extend cgi.FieldStorage
somehow?  I'm sure I just don't have a good enough grasp of what is
going on here to see the light, but I'll spare the group my personal
tutelage.

If anyone has time to pound into my head some conceptual notions of
classes, functions and other fundamentals and has time to write me, I'd
appreciate it. I learn best by doing and getting the basics down pat,
but I seem to be stuck in this example and the book, although great, is
becoming paper which my body is refusing to do anything but stare at.

I can call if in the U.S., and I can pay with beer too ;)

--Rick 
                                ^^^^^^^^^^^^^^
Rick Robino                                  mailto:rrobino at wavedivision.com



From thomas at xs4all.nl  Wed Aug 25 08:08:16 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Wed, 25 Aug 1999 14:08:16 +0200
Subject: Python 1.5.2 on HPUX 10.20
In-Reply-To: <37C3D5BD.72024971@abg1.siemens.de>
References: <37C3D5BD.72024971@abg1.siemens.de>
Message-ID: <19990825140816.R28043@xs4all.nl>

On Wed, Aug 25, 1999 at 01:38:37PM +0200, Volker Hetzer wrote:

> This is my first look into python. I tried to install it
> % ./configure --prefix=/opt/python
> % make all install

> and to test it
> % make test

> The fcntl test failed (twice):

> % /opt/python/bin/python Lib/test/test_fcntl.py
> Status from fnctl with O_NONBLOCK:  0
> struct.pack:  '\000\002\000\000\000\000\000\000\000\000\000\000\000\000\000\000'
> Traceback (innermost last):
>   File "Lib/test/test_fcntl.py", line 30, in ?
>     rv = fcntl.fcntl(f.fileno(), FCNTL.F_SETLKW, lockdata)
> IOError: [Errno 22] Invalid argument
> %

> Is this serious? What am I to do?

You were probably trying to test locking on a file on a filesystem that
doesn't support locking. NFS filesystems, for example, usually dont support
it (unless both your machine and the NFS server have lockd running.)

The test tries to lock /tmp/delete-me by default... If /tmp is a filesystem
that doesn't support locking (it's a RAM-like filesystem on some operating
systems, I am not sure if those support locking) you can change it by
changing the line

filename = '/tmp/delete-me'

to make it point to somewhere that does support locking, and where you can
write (perhaps your homedirectory, though that is often nfs-mounted ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From fredrik at pythonware.com  Tue Aug 17 12:58:15 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 17 Aug 1999 16:58:15 GMT
Subject: Bug in ntpath.exists?
References: <7pbp9c$1b4$1@holly.csv.warwick.ac.uk>
Message-ID: <006601bee8d1$b3b2a1f0$f29b12c2@secret.pythonware.com>

Andrew Clover <esuzm at primrose.csv.warwick.ac.uk> wrote:
> There seems to be a problem in os.path.exists on WinNT, and presumably the
> other os.stat-based calls in ntpath.
> 
> It returns true (well, 1) for objects with any number of spurious '.' and '"'
> characters on the end of a filepath, and any number of spurious spaces on the
> end of the filepath after these. Additionally any path element may contain a
> single spurious '.' character at the end. (Also each path separator may
> consist of any number of forward- or back-slashes, but that's probably
> intended.)

that's an NT file system "feature".  trying to work around
it in the interface layer is probably a bad idea (when you
think you've got it 100% right, Microsoft releases yet an-
other implementation...)

> That is, if the file 'e:\fish\chips.txt' exists, the following paths are also
> claimed by exists to exist:
> 
>   'e:\fish\chips.txt.'
>   'e:\fish.\chips.txt"""'
>   'e:\fish/\/\/\chips.txt....      '

not to mention:

   'e:\FISH\CHIPS.TXT'

;-)

assuming you have an NTFS drive, you can also
try opening "e:\fish\chips.txt::$DATA"...

> This leniency could be considered a good thing in some cases, but not in my
> case since I'm writing a 404 CGI script which attempts to verify the missing
> URL searched for and corrent wrong parts.

to be on the really safe side, your script should
probably use something like:

    fragment in os.listdir(prefix)

to make sure there's really something called fragment
with the given prefix.  (better ideas are welcome...)

</F>




From gerrit.holl at pobox.com  Mon Aug 30 09:03:21 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 30 Aug 1999 13:03:21 GMT
Subject: Parrot: Why the clunky syntax?
In-Reply-To: <37C9FF53.DD466783@compaq.com>; from Greg Ewing on Mon, Aug 30, 1999 at 03:49:39PM +1200
References: <935279776snz@vision25.demon.co.uk> <935454399snz@vision25.demon.co.uk> <37C2B9CE.95AA5FEC@lri.fr> <37C9FF53.DD466783@compaq.com>
Message-ID: <19990830150321.C6858@optiplex.palga.uucp>

On Mon, Aug 30, 1999 at 03:49:39PM +1200, Greg Ewing wrote:
> Instead of a syntax like
> 
>   window @MyWindow "My First Window" {
>        menuBar {
>           menu "File" {
>              menuItem @New "New"
>           }
>   }
> 
> why not something more Python-inspired, e.g.
> 
>   window MyWindow("My First Window"):
>     menubar:
>       menu File:
>         menuitem New
> 
> After all, we like Python because it does without
> line noise like @{} -- don't we?
> 

What are you talking about?
I think I missed something...

regards,
Gerrit.



From tchrist at mox.perl.com  Thu Aug 19 14:47:07 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 11:47:07 -0700
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net>
Message-ID: <37bc431b@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    drew_csillag at geocities.com writes:
:> Thanks for nothing: there are no manpages. 
:Manpages don't work very well if you're stuck on a Windows box :(.  If
:only the world were Unix... <sigh>

So what?  man is a trivial program to write, if all you have
to do is display preformatted text info.  

:There are the python manuals which are excellent, available at:
:http://www.python.org/doc/ (linked to from the homepage)
:available in your choice of HTML, PostSript, PDF, Info, LaTeX and
:windows help.

None of those is text.  I can't use not only man(1), apropos(1),
or whatis(1), I also can't use text-processing tools like grep(1) or
directory processing ones like find(1).  

:Not to mention that they're searchable via:
:http://www.python.org/search/

I wasn't on the Internet when I learnt python.  Please don't shoot me.

:But for the most part, I usually don't need the search portal since the
:indexes are really good.  But if you want a greppable form, there's Info
:(although I don't like info in general, but info (both standalone and
:emacs versions) does have the document search key 's'), HTML, and the
:LaTeX versions grep quite well.

You think HTML greps sell?  I don't think so.  Now, you can argue
that troff doesn't grep well either, and you'd be right.  But
I grep the catpages or podpages.

:No, actually, just trying to help you out, so chill out alright?  

Fine.  I just hate being told to read the fucking manual when there
ISN'T one.  I despite Linux and GNU and the FSF and all that crap
because of this.

--tom
-- 
 "VAX. For those who care enough to steal the very best."
     -- A microscopic message on the silicon chip inside
	one of Digital Equipment's often stolen computer designs.



From dwallace at delanet.com  Sat Aug 28 16:31:35 1999
From: dwallace at delanet.com (Dave Wallace)
Date: Sat, 28 Aug 1999 16:31:35 -0400
Subject: Loading shared modules from an imbedded interpreter
Message-ID: <37C84727.4E1EB9AF@delanet.com>

Hello,

    I am attempting to embed Python in a c executable, unfortunately
when I attempt to import a shared-c module or a Python module that
imports a shared-c module, I get errors like:

  Traceback (innermost last):
  File "<string>", line 1, in ?
ImportError: /usr/local/lib/python1.5/lib-dynload/selectmodule.so:
undefined symbol: _Py_NoneStruct

where the undefined symbol varies from file to file. Apparently, when
compiling parts of libpython are left out since they are not referenced
by my code. Is there any way to force them to be included?

Thanks,
Dave
dwallace at netgsi.com




  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From 71351.356 at compuserve.com  Mon Aug 16 23:49:28 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Mon, 16 Aug 1999 23:49:28 -0400
Subject: stat, chmod and RDONLY file (win32)?
Message-ID: <7pambk$3m8$1@ssauraaa-i-1.production.compuserve.com>

(newbie alert!)
I'm writing a file directory manipulation and system utility executor in
Python, since its supposed to be a good glue scripting language.
(anything, better?)

Hitting a lot of stumbling blocks.   Latest is trying to delete a directory
tree that may have files with a read-only attrib (in NT and win98).      The
exception occurs in shutil.rmtree(), so I guess I'll have to cobble
something to find such files and chg the attributes.   I found os.stat(),
chmod() and stat module, but the docs are a little shallow.   Does the doc
for chmod(): "Change the mode of path to the numeric mode. Availability:
Unix, Windows" mean that one must supply the c-run-time bit flags,
documented elsewhere (where?), as mentioned in the open() API blurb, such as
O_RDONLY.     Sorry, if this is dumb question, lately I *have* been a little
dumb.   There are probably better alternatives to do a 'safe' deltree...

I wrote a little ditty to show file status, maybe?:
def fstat( name) :
   print "%s: [%s]" % (  name,hex( os.stat(name)[ST_MODE] )  )

and get:
locked.txt: [0x8124]
temp.txt: [0x81b6]
----------------------------------------------------------------------------
----------------------------------------------------------------------------
-------------
Hopefully, when this is working and must be used on Unix, there will be no
code changes necessary, right?  : )











From crdempsey at uswest.net  Wed Aug 25 23:51:30 1999
From: crdempsey at uswest.net (Syniq)
Date: Wed, 25 Aug 1999 21:51:30 -0600
Subject: Small question about Python executeable(sp?) files
References: <37C3DAF9.D51DAAF5@uswest.net> <14276.2373.676736.425421@dolphin.mojam.com>
Message-ID: <37C4B9C2.A54444D@uswest.net>

Skip Montanaro wrote:

>
> Check out the various freeze/squeeze/xfreeze
> packages available for Python.  Poking around http://www.python.org/search
> should yield some positive leads.
>
>

Skip and Alex:

Thank you both!  I will check out fr/squ/xfreeze, but I think that's what I
need.  Thank you very much. Thank you, thank you thank you!


I have read in many books and on several websites that you can get help for
__________  (insert random
computer related subject) in comp.something.________   or on some IRC
chatroom...yet when I get there, what I find is either porn or a bunch of
nonsense--no help.  What I've read and what I've experienced with this group
tells me it's different.  Thank you so much for showing me some of the
community support for which linux is so famous.  I hope that at some point,
I will be able to provide the same service to another user.

Syniq




From tim_one at email.msn.com  Mon Aug 23 01:58:25 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 23 Aug 1999 01:58:25 -0400
Subject: UnboundLocalError (was RE: Compound Assignment Operators ( +=, *=, etc...))
In-Reply-To: <37BE7013.C2DA2200@fnmail.com>
Message-ID: <000301beed2c$83b6ece0$352d2399@tim>

[Per Kistler]
> Could you explain to me the meaning of "<wink>", please?
> Is it a form of ":-)"?

It's closer to ;-), but "<0.1 wink>" through "<0.9 wink>" (and an uncountable
number of others) have no literal translation to primitive emoticons.  BTW, I
started doing that when editing a Twin Peaks newsletter that got snail-mailed
to non-computer people as well as emailed to geeks.  The former had no idea
what to make of ;-) and the like, but grokked <wink> & friends instantly.  And
that's why I use Python too.

go-ahead-just-*try*-to-make-a-buck-off-of-that<wink>-ly y'rs  - tim





From Jerome.Kalifa at polytechnique.fr  Mon Aug 16 12:37:05 1999
From: Jerome.Kalifa at polytechnique.fr (Jerome Kalifa)
Date: 16 Aug 1999 18:37:05 +0200
Subject: [NOISE] [OFF-TOPIC] [BUT IMPORTANT] Re: Why use Python when we've got Perl?
References: <000301bee78f$6cd39de0$f22d2399@tim>
Message-ID: <wmc4shzadhq.fsf@balard.polytechnique.fr>

This is absolutely off-topic, definitely pure noise.

"Tim Peters" <tim_one at email.msn.com> writes:

> [Aahz Maruch]
> > ...
> > Arguing over whether Python is "better" than Perl is like arguing
> > over whether European food is better than Asian.
> 
> Exactly so!  Everyone knows Asian food is better, but it's not polite to say
> so in mixed company.
> 

Gasp, shocking! 'nearly died of heart attack when reading that. Nobody
else than Americans would dare writing such other-simplifying
absurdities :-) <--- smiley, don't shoot, we Europeans usually don't
have firearms :-) <--- second smiley.

Indeed, the comparison is silly, but for a different reason : Asian
food, European food... this doesn't mean anything at all! What's in
common between Japanese, Chinese and Cambodian foods, or between
German, English, Spanish and Italian foods? Nothing. BTW, as a biased
arrogant French, I'm not allowed to judge the food of my country, but
of course you know what I think :-)

> gastrowinkingly y'rs  - tim
> 

Veau Marengot rules.
-- 
Jerome Kalifa
Centre de Mathematiques Appliquees,  Ecole Polytechnique.
91128 Palaiseau Cedex, France.  (33)169333981



From mwh21 at cam.ac.uk  Mon Aug 16 12:53:29 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 16 Aug 1999 17:53:29 +0100
Subject: [NOISE] [OFF-TOPIC] [BUT IMPORTANT] Re: Why use Python when we've got Perl?
References: <000301bee78f$6cd39de0$f22d2399@tim> <wmc4shzadhq.fsf@balard.polytechnique.fr>
Message-ID: <m3n1vr3bw6.fsf@atrus.jesus.cam.ac.uk>

Jerome Kalifa <Jerome.Kalifa at polytechnique.fr> writes:
[schnupp]
> Indeed, the comparison is silly, but for a different reason : Asian
> food, European food... this doesn't mean anything at all! What's in
> common between Japanese, Chinese and Cambodian foods, or between
> German, English, Spanish and Italian foods? Nothing. 

Well, they're all carbon based, genereally.

The funniest part of the perl/Python flame war is how similar
languages are, especially when compared with stuff like lisp, ml or
haskell.

oh-my-god-I'm-about-to-post-to-a-perl-vs.-Python-thread-ly y'rs
Michael



From amaranda at nospam.com  Tue Aug 31 05:06:24 1999
From: amaranda at nospam.com (Alex Maranda)
Date: Tue, 31 Aug 1999 10:06:24 +0100
Subject: All rise for judge Timothy Peters...
References: <slrn7sloo9.r4u.init__self@jules.com>
Message-ID: <37CB9B10.DF67F398@nospam.com>

Python Addict wrote:
> 
> I kid you not. I'm involved in a legal battle (on the prosecution
> side, thank you very much) and spent a morning in court just
> recently. My flabbers where ghasted when I noticed the Judge's name
> plate.
> 
> Now "our" Tim always goes at length to not explain what he does so
> I'm wondering if these two are the same person...
Can't be him. He's only a bot, remember?
<bots-can't-have-bodies-even-the-damn-good-ones'ly-yours> alex



From amaranda at spider.com  Mon Aug 23 09:48:24 1999
From: amaranda at spider.com (Alex Maranda)
Date: Mon, 23 Aug 1999 14:48:24 +0100
Subject: How to sell Python to your boss?
References: <ka8yaf7d84t.fsf@delightful.cs.hut.fi> <wjywvur2qrn.fsf@bradh.ne.mediaone.net> <37BD2F89.6CC2@spider.com> <659B7DD56CD4A360.51878E54B911E668.D1EDB31DC30BE6DF@lp.airnews.net>
Message-ID: <37C15128.6598@spider.com>

Cameron Laird wrote:
>                         .
> Sometimes Python is a good answer when one *is* stuck
> with "a pile of legacy code".  That is, Python is quite
> adept at "gluing", or "wrapping".  When you find your-
> self with an intractable application that needs updating,
> one of your first reactions should be, "What would happen
> if I could access some of this functionality through a
> Python interface?"
Aye, but there are legacies and legacies. This of the fun one has when
the aforementioned crap goes into kernel space.

Cheers,
-- 
Alex Maranda         mailto: amaranda at spider dot com
Spider Software Ltd. Tel: +44 (0)131 4757036
Edinburgh, UK        http://members.xoom.com/Alex_Maranda
STREAMS based communications protocols for embedded systems



From xah at best.com  Tue Aug 24 18:15:42 1999
From: xah at best.com (Xah)
Date: Tue, 24 Aug 1999 15:15:42 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk>
	 <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu>
	 <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
	 <37b4d0e0@cs.colorado.edu> <37b6500c.2143627@read.news.global.net.uk>
Message-ID: <37c3198f$0$215@nntp1.ba.best.com>

In article <37b6500c.2143627 at read.news.global.net.uk>, 
dave at see.sig.for.address (Dave) wrote:
>
> John, please refrain from cross-posting all your messages to
> comp.lang.python. If there is anything worse than getting a long and
> pointless flame war, it is getting half a flame war. By cross posting
> you are clearly trying to drag the readers of c.l.py into the
> argument, which is pure trolling.  If you continue you will merely
> succeed in pissing off two newsgroups instead of one.

In geek religion wars, there are two types of people. One of them are the
nincompoops, who try to do everything in their power to advocate their
superiority. The other is the super-nincompoops, who, instead of sitting
their asses on high clouds, have to dive in to make an announcement that the
nincompoops' behaviors are inappropriate.

I have a message for the super-nincompoops class: "shut ya pie hole".

> The 'my language is better than yours' debates are always pointless
> and counterproductive, and perl vs. python has been done countless
> times before. Life is too short, live and let live [insert other
> cliches to taste].

Perhaps YOUR life is too short, not mine! I have thoroughly enjoyed the
intricate discussions, and Perl is the language that will save the universe
from collapsing.

> Incidentally you are wrong about perl arrays and python lists being
> different data structures - internally they are both implemented as C
> arrays, only the names are different.

Incidentally really? Gather your brain cells and let's see if you can pull
off some real content. You might magically get out of my killfile.

>   Dave Kirby
>
> --------------------------------------------------
> All great ideas start as heresy and end as dogma.
>
> dkirby@   <-figure this out, spambots!
> bigfoot.   My opinions are my own,
> com        but I'm willing to share.

 Xah
 best.com at xah    <- figure this out, donkey!
 http://www.best.com/~xah/PageTwo_dir/more.html
 "All stupid sayings start and end as tomfool's quotes."



From bwarsaw at cnri.reston.va.us  Thu Aug 19 14:56:57 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 14:56:57 -0400 (EDT)
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
	<m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <14268.21369.636710.682732@anthem.cnri.reston.va.us>

    MH> At some point Python will probably support `extended slice
    MH> syntax' like so:

    MH> range(0,10)[1:4:2] => [1,3]

    MH> Numerical Python already does this.

As does JPython:

JPython 1.0.3 on java1.2
Copyright 1997-1998 Corporation for National Research Initiatives
>>> range(0,10)[1:4:2]
[1, 3]

Not surprising given the author of NumPy and JPython :)

-Barry



From philh at vision25.demon.co.uk  Sun Aug  8 09:11:14 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 08 Aug 99 13:11:14 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <19990805094020.A1838753@vislab.epa.gov> <37aacdf9.324492916@news.oh.verio.com>
Message-ID: <934117874snz@vision25.demon.co.uk>

In article <37aacdf9.324492916 at news.oh.verio.com>
           ap096 at po.cwru.edu "Kevin Dahlhausen" writes:
> Under linux, there is a wrapper for fltk that translates the data files from the> FLTK gui builder to Python.  This one is very much a work in progress though.
> It should also work under Windows, but that is still theory at this point.
> http://fltk.netpedia.net

How does fltk compare with other GUI libraries such as Tk, GTK, KDE/Qt,
etc? Is there room for yet another user interface library?
 
> Randall Hopper <aa8vb at yahoo.com> wrote:
> > |Is there a Python GUI builder out there anywhere?

I'm currently writing a generic GUI builder.

It doesn't (and won't) have a graphical user interface, the user
will write a user interface definition file, which will then be
compiled by my tool into the programming language and GUI toolkit of
your choice. A definition file might look like this:


table align=right {
   tr {
      td { label "Name:" }
      td { textField @Name cols=10 }
   }
   tr {
      td { label "Type:" }
      td { radioButton @Int "int" 
           radioButton @String "string" 
           radioButton @Float "floating point"
      }
   }
   tr {
      td { label "Description:" }
      td { textField @Desc cols=20 }
   }
}


Which might produce a form like this:

+-----------------------------------------------------------+
|  Name:         [..........]                               |
|                                                           |
|  Type:         (*) int  ( ) string  ( ) floating point    |
|                                                           |
|  Description:  [....................]                     |
+-----------------------------------------------------------+


The program is written in Python.
At the moment, I'm working on producing Python/Tkinter and HTML
back ends. I'm trying to make the back ends modular from the
main program and easy to code.

The idea is that it will, in principle, be able to generate code for 
any back end. (Once someone has written that back end, of course).
The advantage of this is that the UI definition language only needs 
to be learned once, and can be used for lots of languages and GUI
toolkits (of which there are a lot around).


-- 
Phil Hunt....philh at vision25.demon.co.uk




From frehe491 at student.liu.se  Sat Aug 21 07:42:17 1999
From: frehe491 at student.liu.se (Fredrik Henbjork)
Date: Sat, 21 Aug 1999 13:42:17 +0200
Subject: What is a troll by the way? [offtopic]
References: <HSAATIAIVmv3Ewd1@jessikat.demon.co.uk>
Message-ID: <99082113432403.12112@o112>

On Sat, 21 Aug 1999, you wrote:
> In article <37BDF3A7.75FEA04F at vision.arc.nasa.gov>, Chad Netzer
> <chad at vision.arc.nasa.gov> writes
> >Robert Kern wrote:
> >
> >> _Troll_ is a USENET term for someone that deliberately posts an
> >> inflammatory message simply to stir up argument.
> >
> >I've found it funny to watch this term evolve over the past few years.
> >It started (I believe) as a reference to the fishing term, and used to
> >mean "trolling for flames" (ie. hoping to bring flames to the
> >discussion), perhaps employing "flame bait" for this purpose.
> >
> ...
> ppresumably the fishing reference is a mis-pronounced form of trawl
> although given the number of English dialects troll might be correct
> somewhere in that sense.
> -- 
> Robin Becker

I think you are wrong. I was an active fisher until I gave it up some years ago
due to animal rights reasons. Trolling and trawling are totally different things.
Trawling means that you basically drag a *big* net behind a *ship* while
trolling means that you have one or several fishingrods with baits trailing
some tens of meters behind the boat. Trawling is industrial fishing done by
professional fisherman while trolling is mostly done by people as a hobby.
Trolling also makes more sense in a semantical way in the expression
"trolling for flames" than trawling since you don't use a bait (remember
"flamebait") when trawling.

--
Fredrik Henbjork

Email: frehe491 at student.liu.se
WWW: http://o112.ryd.student.liu.se



From m.faassen at vet.uu.nl  Sat Aug 28 13:58:41 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Aug 1999 17:58:41 GMT
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908271102120.14054-100000@eris.io.com>
Message-ID: <7q980h$7bi$2@newshost.accu.uu.nl>

Bit Nik <bitwise at io.com> wrote:

> I just recently joined this newsgroup/mailinglist and I missed out a lot
> on the previous flame war (lucky me!). And I certainly don't want to
> stroke the coals and recreate the flame, but i'm very curious /how/ is
> Python lacking in documentation? This seems to be the major argument
> against it right?

Oh, that was just a silly flamewar over Python having no man pages of the
library reference. I think all parties involved agreed in the end that it
got a bit out of hand.

> Also, who are these Python biggots people are always
> talking about??

[snips]
> Another quick question, are there such things as PUGs? (Python User
> Groups). I couldn't find any information on python.org...

I believe you have various concepts confused. First of all, I must note
here that Dutch is the Secret Natural Language of Python. Don't ask me
why, it's just a cosmic conspiracy that all Pythoneers just happen to speak
Dutch.

Then, I must add that 'big' in Dutch means 'young pig' in English. 
Also I must point out that PUGs are not called PUG but PIG, for 
'Python Interest Group'. There's an underlying pattern emerging here, as you
can see. According to the Python Homepage there are DCPIGgies and PORPIGies
in Washington DC and Portland respectively. For Genuine Python biggots 
therefore, go to those piggies. :)

It's-actually-spelled-bigots-ly yours,

Martijn

P.S. Seriously, I'm just being silly. Welcome to the Python community!




From ron at library.ucsf.edu  Mon Aug 23 15:39:26 1999
From: ron at library.ucsf.edu (Ron Klatchko)
Date: Mon, 23 Aug 1999 12:39:26 -0700
Subject: Sending a mail
References: <37C1704B.D1B173E5@daimi.au.dk>
Message-ID: <37C1A36E.7DC4653@library.ucsf.edu>

Peter Posselt Vestergaard wrote:
> #!/usr/local/bin/python
> import os
> text='echo "hello"|mailto posselt at daimi.au.dk -s "important message"'
> os.system(text)
> 
> This works quite fine in my python interpreter, but not then I put it on
> the web. I tried to print out the standardoutput from os.system(text)
> and it was 256. Probably my python-server won't allow me to execute that
> command or what? 

Although it's possible that mailto cannot be run by a CGI, before I went
that route I would try using an absolute path for mailto in your command
line.  CGIs tend to have limited environments so it probably doesn't
have the path to mailto.

moo
----------------------------------------------------------------------
          Ron Klatchko - Manager, Advanced Technology Group           
           UCSF Library and Center for Knowledge Management           
                        ron at library.ucsf.edu



From ovidiu at cup.hp.com  Mon Aug 16 20:43:27 1999
From: ovidiu at cup.hp.com (Ovidiu Predescu)
Date: Mon, 16 Aug 1999 17:43:27 -0700
Subject: function which knows who has called it. 
In-Reply-To: Your message of "Mon, 16 Aug 1999 22:12:49 GMT."
             <934841569snz@vision25.demon.co.uk> 
Message-ID: <199908170043.RAA32717@hpcll563.cup.hp.com>

On Mon, 16 Aug 99 22:12:49 GMT, philh at vision25.demon.co.uk (Phil Hunt) wrote:

> Is if possible in Python to have a function that prints the name
> of the function that called it. (I'm thinking of using this in
> some debugging code I want to write).

Yes, it's possible, take a look at the traceback module. I use this piece of
code to print the name of function that invokes the 'should' function:

import traceback

def should (condition, message=None):
    if (not condition):
        methodName = traceback.extract_stack()[-2][2]
        fail ("%s: %s" % (methodName, message))


Greetings,
-- 
Ovidiu Predescu <ovidiu at cup.hp.com>
http://andromeda.cup.hp.com/  (inside HP's firewall only)
http://www.geocities.com/SiliconValley/Monitor/7464/





From fdrake at acm.org  Wed Aug 18 13:28:53 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Wed, 18 Aug 1999 17:28:53 GMT
Subject: representation of unbound methods - strange behaviour
In-Reply-To: <37bbecf1.100354231@news.omnilink.de>
References: <37bbecf1.100354231@news.omnilink.de>
Message-ID: <14266.60757.673964.501369@weyr.cnri.reston.va.us>

Stefan Franke writes:
 > Why does this happen:
...
 > <unbound method A.f>
 >                 ^^^

  The "f" part of the name is taken from the function object, not the
name used to access it:

     >>> class B: pass
     >>> def h(): pass
     >>> B.f = B.g = h
     >>> B.f
     <unbound method B.h>
     >>> B.g
     <unbound method B.h>

  Whether this should be this way is open to debate, but the cost of
doing it differently probably involves doing more work in creating the 
unbound method object, which usually gets thrown away quickly.  Since
you don't normally want the repr of it at all, it's probably not worth 
changing, since that would likely slow down all method invocations.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From mal at lemburg.com  Tue Aug 10 03:32:57 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 10 Aug 1999 09:32:57 +0200
Subject: RPM for Python 1.5.2 ?
References: <slrn7qrn3u.7fj.dyoung@baltazar.baltazar.org> <37ADE4C1.194C31D6@lemburg.com> <19990809113731.A6757@baltazar.baltazar.org>
Message-ID: <37AFD5A9.4C45332F@lemburg.com>

David Young wrote:
> 
> It says at that URL that the binaries are compiled with pgcc with
> -mpentiumpro, etc. It's essential that they work for me on any
> 386-compatible architecture. Will they? If not, do you know of a package
> that will?
> 
> Dave
> 
> On Sun, Aug 08, 1999 at 10:12:49PM +0200, M.-A. Lemburg wrote:
> > David Young wrote:
> > >
> > > Has anyone built an RPM for Python 1.5.2 ? I think it's sort of curious
> > > that RedHat hasn't already. Anyway, I'm building my own package right now,
> > > but I'm afraid I might be wasting my time....
> >
> > Try:
> >
> >       http://www.andrich.net/python/download.html

They should work on 386 machines too -- but slower than ones compiled
for i386.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   143 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From joelg at alum.mit.edu  Mon Aug  9 18:44:29 1999
From: joelg at alum.mit.edu (Joel Gould)
Date: Mon, 09 Aug 1999 22:44:29 GMT
Subject: NatLink
References: <7oletd$jdj$1@nnrp1.deja.com>
Message-ID: <37af5918.10124167@news.ne.mediaone.net>

Someone has started a mailing list for users of NatLink.  The address
of the mailing list is:

http://harvee.billerica.ma.us/mailman/listinfo/natpython

The list is not very active (yet), but I have been known to answer
questions posted to this mailing list.

Joel Gould
Creator of NatLink

On Mon, 09 Aug 1999 02:37:34 GMT, infody864 at my-deja.com wrote:

>I would be interested in conversing with programmers for using Python
>and NatLink.
>If anyone is interested in helping each other, please contact me.




From robin at jessikat.demon.co.uk  Thu Aug 19 13:58:08 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 19 Aug 1999 18:58:08 +0100
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
 <slrn7ro9ns.h6v.jblaine@shell2.shore.net> <37bc2ee5@cs.colorado.edu>
Message-ID: <nJUYxAAwWEv3EwY0@jessikat.demon.co.uk>

In article <37bc2ee5 at cs.colorado.edu>, Tom Christiansen
<tchrist at mox.perl.com> writes
...
It's a good thing I'm in the killfile as otherwise my response would be
raising the blood pressure of this arrogant person. Perl with any amount
of documentation sucks when compared to python. It's not so much the bad
language structure/syntax but the lack of sensible advice. Also (it's a
number thing) there are good reasons for maths/engineering types like
myself to prefer python. 
>
>Don't be silly.  That wasn't that at all.  Online searchable documentation
>is critical in a technical subject like this.  Imagine libc without
>manpages!  Or having no manpages for the shell tools. Now, you might
>now be aware of this, but I happened to *write* most of that on-line
>documenation for Perl.  I'm super senstitive to not being able to look
>up docs using standard CLI text-based tools.  As you said, you would be
>frustrated not to have docs for tools or libraries.
>
>Something the python community might should learn here from the perl
>community is our zealous position on usable documentation.  HTML alone
>sucks, as does putting things only on the web (two different issues).
>That I can't even get "man python" to come up pisses me off more than
>you can imagine.  I strongly feel that you really should consider making
>something easy for CLI people to use just like any other tool or library.
>That's why we have "pod" as base docs, which can convert into HTML,
>text, info, latex, frame, troff (manpages), etc.  It makes it easy to
>run grep or find programs on.  This is very important.  And each module
>automatically installs that module's manpage (and optionally html doc,
>but this is less popular amongst perl folks).  That gets it added to
>the apropos and whatis and man commands automatically.  This is about
>0.002% of what the MakeMaker thingie in Perl does, but it's the part
>that matters for docs.  I've surely heard Python folks drooling over it
>for a long time.  Also, the CPAN autoinstaller than runs it all for you.
>All very spiffy.
>
>--tom

-- 
Robin Becker



From gmcm at hypernet.com  Tue Aug 31 14:38:56 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 31 Aug 1999 13:38:56 -0500
Subject: C++ and Python
In-Reply-To: <7qgnr5$ls8@news.go.dlr.de>
Message-ID: <1276001624-57794109@hypernet.com>

Sven Drescher writes:
> 
> I have a little problem. It's really urgent. I wrote a program with
> threads. One thread is the Python/Tk loop with my graphical user
> interface and the other is an work thread.
> 
> Now my question:
> How can I get the entrie-variables of python in my work thread.
> 
> I know such functions like PyArg_ParseTuple and so on to convert the
> values, but is there a possibility to have an C-reference variable
> to a python variable or so?

Short answer: "no".

Problem 1:  Difference between reference semantics and slot 
semantics.

 foo = "I'm a string"
 foo = foo + ", but now I'm longer"

In C, the first foo has no relationship to the second. In a case like 
this (references to immutables where the reference may change), you 
have to look up "foo" in the appropriate dictionary.

Problem 2: scope. What dictionary does it live in?

Well, I think we can assume you're only talking about globals. In 
which case, the longer answer is "yes, sort of". You can of a 
module's dictionary in your C code and look things up in it. You 
could even optimize this (only look it up once) for those cases where 
you know the reference doesn't change, but that's a bit risky:

 a = []
 a.append(3) # reference doesn't change
 a = a[:] # oops, reference changed.

As long as you don't release the interpreter lock in your C code, you 
won't have to worry much about race conditions (everything you do 
will be atomic as far as Python is concerned). If you do release the 
interpreter lock, make sure any Python variables are in a consistent 
state, and don't touch them until you've re-aquired the lock.

- Gordon



From chad at vision.arc.nasa.gov  Sun Aug 29 15:29:50 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Sun, 29 Aug 1999 19:29:50 +0000
Subject: Returning none
References: <37C55044.8CA02530@prescod.net>
			<37C5683D.91D61FD0@rubic.com>
			<14277.28442.44107.591412@dolphin.mojam.com>
			<37C70821.92E810FC@prescod.net> <14279.9024.401900.142461@dolphin.mojam.com> <37C7B86E.647BA654@prescod.net>
Message-ID: <37C98A2E.3F21A7E@vision.arc.nasa.gov>

Paul Prescod wrote:

> Makefile? Isn't that what I'm using Python to get away from?

Well, you could just write a script that first runs PyLint, then runs
your program (it could even get fancy and check for errors and not
run it if there were any, and store the PyLint output for perusal, etc.)

You could also make this part of your python program; when it starts
up, PyLint itself, etc.

If I weren't busy, I'd even attach such a script.  Anyone else?

Chad Netzer
chad at vision.arc.nasa.gov






From I.Clarke at NOSPAM.strs.co.uk  Thu Aug  5 04:39:20 1999
From: I.Clarke at NOSPAM.strs.co.uk (Ian Clarke)
Date: Thu, 05 Aug 1999 09:39:20 +0100
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com>
Message-ID: <37A94DB8.6C9617C3@NOSPAM.strs.co.uk>

> I'm not going to say either of those! I have done it, although I'm not
> terribly happy with the results so far and it's on the back burner for
> the moment .... I need to figure out if it's still useful in a Zope
> world - which means I have to learn and understand Zope!

Yes, I am in the same situation.  It sounds from a user interface point
of view, that Zope can do Python embedding easily, so then the question
becomes whether this idea would be more efficient than Zope (in terms of
page generation, AND simplicity of deployment).

> I embed the python fragments in HTML comment blocks for the sake of
> my HTML editor tools which would otherwise feel very sad ...

Well, I wanted to allow the editing of Python code blocks from within
the HTML editor (that would prevent any kind of autoindentation).
 
> I've also set it up (at least in one version) so that the python can
> generate http headers at the start as well .... or they can be embedded in
> the templates.

Cool!

> Yup! All there (minus "appropriately neat": more "rather nasty" for now I'm
> afraid)

I thought that the CGI variables could be directly translated into local
variables and accessed like that, or possibly just turned into a
dictionary called 'cgi'.

> Also I haven't yet dealt with putting the python into rexec for execution-
> but this is only needed if a web master wants to give their clients access
> to these facilities as opposed to doing things themselves.

I think this would be a very nice feature as it would mean that ISPs
could grant CGI access to their users (provided their users are willing
to learn Python - and what language could be easier to learn?) without
raising any security concerns.

> * indenting of the python bits versus the HTML bits doesn't mesh
>    very well

Yes.  We look into writing indentation modes for Emacs, and combining
HTML and Python modes in some way.
 
> * the base server that I use is a threaded version of SimpleHTTP and I seem
>    to have occasional problems with the thing hanging ... don't know why
>    and haven't time to look into it just now

Peculiar, if the problem could be replicated then we could get whoever
wrote SimpleHTTP on the case.

> * I haven't figured how many security holes this might open ...

Hopefully none if we execute the Python scripts in a "padded cell".

> I'll upload the thing to my starship account when I get home tonight just
> in case anyone is interested!! I'll get back tomorrow with a URL.

I look forward to taking a look.  Are you willing to release the code
under the GPL?

Ian.

-- 
Remove the "NOSPAM." from my email address before replying.



From tchrist at jhereg.perl.com  Thu Aug 19 12:34:12 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Thu, 19 Aug 1999 16:34:12 GMT
Subject: More random python observations from a perl programmer 
In-Reply-To: Andrew Csillag's message of Thu, 19 Aug 1999 16:31:02 -0400 .
             <37BC6986.A66B044E@starmedia.net> 
References: <37BC6986.A66B044E@starmedia.net>
Message-ID: <199908191634.KAA01333@jhereg.perl.com>

>> How is that a singleton tuple? 
>It's not.  For format strings, if the item (here 4.5/100) is a single
>thing, the items need not be a tuple. i.e.

That's painfully inconsistent.

>>     print "You have $%0.2f bucks" % (4.5/100,)
>> illegal, 
>No, it's perfectly fine.  Didn't try it, I guess.

No, I thought I had.

>>     print "You have $%0.2f bucks" % ((4.5/100,) * 2)
>This, on the other hand is wrong since ((4.5/100,) * 2) yields (0.045,
>0.045) and you only have one formatting bit in the string (only 1
>%0.2f).  

Typo.

>> mandatory?  Was this (elt,) thing just an, er, unforeseen design
>> misfeature? Yes, I tripped me up again. :-(
>Try reading the docs...

Thanks for nothing: there are no manpages.  I read the Lutz book.
You'll note that I even MENTIONED this issue in a gotcha.
It's just easy to get tripped up by the inconsistency.

You seem to be looking for a fight.   Please say nothing
at all if that's all you want to do.

--tom



From jeremy at cnri.reston.va.us  Thu Aug 19 14:28:51 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 19 Aug 1999 14:28:51 -0400 (EDT)
Subject: More random python observations from a perl programmer 
In-Reply-To: <199908191819.MAA04074@jhereg.perl.com>
References: <14268.19155.505284.821500@bitdiddle.cnri.reston.va.us>
	<199908191819.MAA04074@jhereg.perl.com>
Message-ID: <14268.19683.897141.770777@bitdiddle.cnri.reston.va.us>

Heck, in the postmodern world of programming languages it probably
means neither and both.  

Jeremy



From hardwork at freemail.c3.hu  Sat Aug 21 21:58:46 1999
From: hardwork at freemail.c3.hu (hardwork at freemail.c3.hu)
Date: Sun, 22 Aug 1999 01:58:46 GMT
Subject: RFC: Viper: yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au>
Message-ID: <7pnlgi$fli$1@nnrp1.deja.com>

In article <37b075ae.2367293988 at news.triode.net.au>,  skaller at maxtal.com.au
wrote:

> This is a request for comments on Viper, Yet Another implementation
> of Python. I'm building Viperi, the interpreter,  and I'm interested in
> comments on three issues:
>
> 	1) compatability with Python
> 	2) Extensions
> 	3) Optimisation
>
> Viperi is being built as a prelude to implementation of an optimising
> compiler, Viperc. Unlike previous attempts at a compiler, Viperc will do
> type inference, inlining, and special case analysis,
> in an attempt to optimise generated code.


The discussion of Viperi and Viperc is interesting.

I would like to know if there are any url available for the rest of us to get
more information regarding either or both of viperi and viperc?

Thanks in advance.


Sincerely,
Bob
hardwork at freemail.c3.hu


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From m.faassen at vet.uu.nl  Sat Aug 28 08:19:34 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Aug 1999 12:19:34 GMT
Subject: Why use Perl when we've got Python?!
References: <000801bef12c$fef6e140$ad2d153f@tim>
Message-ID: <7q8k4m$pua$1@newshost.accu.uu.nl>

Tim Peters <tim_one at email.msn.com> wrote:
> [Kent Polk]
>> I believe you are thinking of Al'Khwarizmi, who wrote the book
>> "Hisab al-jabr w'al-muqabala". 'Algebra' came from 'al-jabr' and
>> 'Algorithm' came from the name Al'Khwarizmi.

> Nonsense.  Everyone knows 'Algorithm' came from Al Gore, Vice President of the
> United States and Inventor of the Internet.

*falls down on the floor laughing*

>> Later Greeks certainly appear to have used the concept of
>> variables, but did have not the algebraic notation.

> Well, at least you got *that* part right <wink>.

> and-john-backus-is-the-god-of-wine-or-something-like-that-ly y'rs  - tim

and-boole-has-something-to-do-with-that-french-game-with-the-balls-ly yours,

Martijn




From MHammond at skippinet.com.au  Fri Aug  6 20:01:40 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Sat, 7 Aug 1999 10:01:40 +1000
Subject: Python in web page
References: <37a70f28.96313821@midpoint> <7o7ta9$roe$1@m2.c2.telstra-mm.net.au> <7oeopb$i1k$1@ns.mtu.ru>
Message-ID: <7oft25$3nm$1@m2.c2.telstra-mm.net.au>

Oleg Orlov wrote in message <7oeopb$i1k$1 at ns.mtu.ru>...
>> Under Microsoft Internet Explorer you can use Python as an Active
>Scripting
>> language.
>>
>
>But python will serialize web clients due to global locking object in
>pytnon.dll.
>May be i missed something?

Yes you did miss something - it will not "serialize" clients at all.

Firstly, the reference was to IE, where there really arent "multiple
clients" - but there may be multiple windows.

Secondly, the internal Python lock is well documented, and it unlikely to
cause performance problems unless you are running CPU intensive code in
Python on a multiple CPU box.  Most IE users are running single CPU users
but even so, Python is rarely doing CPU intensive work - it is typically
waiting on something external, such as the user, network, files, database or
mail server, etc.

Mark.






From bwarsaw at cnri.reston.va.us  Thu Aug 19 10:52:55 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 10:52:55 -0400 (EDT)
Subject: whrandom documentation (was Re: Unsorting(randomizing) a sequence)
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<14265.47748.241386.291144@buffalo.fnal.gov>
	<wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
	<m3672dwojr.fsf@pcrm.win.tue.nl>
	<14266.49377.51221.305982@dolphin.mojam.com>
	<14266.59802.56315.731454@bitdiddle.cnri.reston.va.us>
Message-ID: <14268.6727.737269.35392@anthem.cnri.reston.va.us>

>>>>> "JH" == Jeremy Hylton <jeremy at cnri.reston.va.us> writes:

    JH> The intended use of the standard library here is for users to
    JH> import random to get access to whrandom.  whrandom implements
    JH> a specific method for generating random numbers.  It could
    JH> easily be replaced with a different, improved module at some
    JH> time in the future.  If user code imports random, it will
    JH> automatically get the benefit of the new implementation.

In that case, shouldn't the documentation for the whrandom module
recommend this usage?

    http://www.python.org/doc/current/lib/module-whrandom.html

-Barry



From g_v_wilson at my-deja.com  Tue Aug 24 22:06:45 1999
From: g_v_wilson at my-deja.com (g_v_wilson at my-deja.com)
Date: Wed, 25 Aug 1999 02:06:45 GMT
Subject: problem compiling extension on Win NT
Message-ID: <7pvj3d$u5m$1@nnrp1.deja.com>

I am trying to build a small Python extension
in C on Windows NT using the Microsoft 'cl'
compiler from the command line (long story).
Everything is working, except my attempts to
#define USE_DL_EXPORT and then use DL_EXPORT.
I have tried #defining the former both before
and after #including Python.h, but DL_EXPORT's
expansion is never correct.  Has anyone run
into this?  If so, I'd be grateful for help...

Please reply directly to:

 g v w i l s o n @ i n t e r l o g . c o m

as I am not a regular reader of this group.

Thanks,
Greg

p.s. anyone out there building Fortran extensions
on Windows with G77?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kistler at fnmail.com  Sat Aug 21 11:35:49 1999
From: kistler at fnmail.com (Per Kistler)
Date: Sat, 21 Aug 1999 17:35:49 +0200
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com>
Message-ID: <37BEC755.645A350D@fnmail.com>

Hi

...but Tom has been attacked over and over again. I can really
understand that he may loose temper after beeing treeted like
a idiot again and again. And Tom is really not any dummy. 
I suggest to treat the "bigger" guys in the computer community
with more respect, which would automatically prevent any "discurse"
like the preceeding one. If that's not possible, I suggest another
way: Zen-meditation. This adds a exception handling wrapper
around the language-syntax-torn mind:-)

Cheers, Per.

srenner at lycosmail.com wrote:

> The reactions Mr. Christiansen evokes may have more to do with his style
> than his content. It strikes me as pompous, pretentious, hostile,
> autodidactic, preening, vain and unpleasant. Perhaps this is not the
> right newsgroup for him.


-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From fdrake at acm.org  Thu Aug 19 14:48:51 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 19 Aug 1999 18:48:51 GMT
Subject: Writing XML from Python ...
In-Reply-To: <m3iu6ccl8m.fsf@intergate.bc.ca>
References: <m3iu6ccl8m.fsf@intergate.bc.ca>
Message-ID: <14268.20883.995845.418264@weyr.cnri.reston.va.us>

Gerald Gutierrez writes:
 > I've been having an extremely hard time finding information on how I
 > can write out XML files from Python. The python-xml software seems to
 > be targetted primarily at parsing XML files, not creating them.

  If you create DOM objects using the XML package (which can be done
programmatically without a source document), you can simply write the
string returned by the document object's toxml() method.
  If you don't want to use the XML package, print works just fine.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From bowman at montana.com  Sun Aug  8 22:09:15 1999
From: bowman at montana.com (bowman)
Date: Sun, 08 Aug 1999 20:09:15 -0600
Subject: RPM for Python 1.5.2 ?
References: <slrn7qrn3u.7fj.dyoung@baltazar.baltazar.org>
Message-ID: <37AE384B.2B5CA412@montana.com>

David Young wrote:
> 
> Has anyone built an RPM for Python 1.5.2 ? I think it's sort of curious
> that RedHat hasn't already. Anyway, I'm building my own package right now,
> but I'm afraid I might be wasting my time....

I didn't do a rpm for RH5.2, just built from source. Installing to
/usr/local/bin and /usr/local/lib tends to break glint, setup, and a
couple of other Python based rh tools, but you can move their .so stuff
and rand.py to the appropriate directories and get the tools working
again. 

I had done the rpm earlier, but vim5.4 didn't seem to find a python
config it needed, so I was having trouble getting the python features
enabled.

After spending a day trying to get PerlTk up. only to discover Perl is
broken in rh5.2, and a few other RPM gotchas, I'm back to building from
source. ymmv

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From zontar4 at yahoo.com  Thu Aug 26 04:34:21 1999
From: zontar4 at yahoo.com (G. Lewis)
Date: Thu, 26 Aug 1999 08:34:21 GMT
Subject: idling with tkinter
Message-ID: <7q2u6e$aso$1@nnrp1.deja.com>

I've been trying to figure out how to hook into idle processes
with tkinter. It's true, i don't know what i'm doing. Could
someone share or point to some source code which might provide
a simple demonstration of this: say, for example, just putting
a 'clock' in a window and keeping it constantly updated... or
at least occassionally updated.

I found the after_idle() method, but it only calls the callback
once. Strangely i can't find a method to perminently hook a
callback into the idle system. Is there one?

--
...


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From philh at vision25.demon.co.uk  Mon Aug 23 21:04:12 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 24 Aug 99 01:04:12 GMT
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan>
Message-ID: <935456652snz@vision25.demon.co.uk>

In article <hhemgu41z7.fsf at forcix.roof.lan> forcer at mindless.com "forcer" writes:
> Hi there.
> 
> Lately, i've been thinking about Python. While it is an
> exceptionally clean language, it has one very obvious wart:
> The ``pass'' statement.
> 
> To quote the reference manual:
> 
>    pass is a null operation -- when it is executed, nothing
>    happens. It is useful as a placeholder when a statement is
>    required syntactically, but no code needs to be executed, for
>    example:
> 
>    def f(arg): pass    # a function that does nothing (yet)
> 
> So it happens that Python has places where semantically nothing
> should be done, while syntactically it requires a statement.
> 
> I consider this suboptimal.
> 
> As a solution, one could allow the following:
> instead of
> 
>         if foo:
>            pass
>         bar
> 
> one could write
> 
>         if foo:
>         bar
> 
> Syntactically this does not introduce any ambiguity. The only
> problem i see is that the "pass" version is more readable.
> 
> But...
> Most of the time, people write
> 
>         if foo:pass
>         bar
> 
> anyways so the "readability" does not suffer from leaving pass
> out.
> 
> Also, the requirement for pass makes it impractical to comment
> out parts of an if expression, like in
> 
>         if foo:
>            # baz
>            # qux
>         bar
> 
> The new syntax would even allow a smooth transition, or
> coexistence of both versions. Pass can be retained to keep the
> possibility for a explicit do-nothing statment, while it's just
> not mandatory anymore.
> 
> I'd like to see such a change in the Python language, so it'll be
> even cleaner than it was before.
> 
> I hope my opinions and thoughts are of value to you,
>        --forcer

<aol>I agree with all this</aol>

-- 
Phil Hunt....philh at vision25.demon.co.uk




From johnm at magnet.com  Tue Aug 24 14:55:48 1999
From: johnm at magnet.com (John Mitchell)
Date: Tue, 24 Aug 1999 18:55:48 GMT
Subject: Necessity of ``pass''
In-Reply-To: <vndaerht6tv.fsf@betelgeuse.ccs.neu.edu>
References: <vndaerht6tv.fsf@betelgeuse.ccs.neu.edu>
Message-ID: <Pine.SGI.4.05.9908241451460.5499-100000@lemur.magnet.com>

On 24 Aug 1999, Justin Sheehy wrote:

> forcer <forcer at mindless.com> writes:
> 
> > You didn't quote the example where i pointed out the real
> > problem with pass:
> > 
> > if foo:
> >    # bar
> >    # baz
> > qux


Try this:

if 0 and foo:
	bar
	baz
qux


If you enable/disable this code often, try this one:

if 01 and foo:
	bar
	baz
qux


A one-character change, deleting/removing the '1', will disable/re-enable
the code. (I first heard this technique from Roy Edwards for C code.)


And for the record, I use "pass" quite a bit.  It's just a marker saying
loud and clear "do nothing", as opposed to the very-shakey ";" do-nothing
in C.


- j





From graham at sloth.math.uga.edu  Mon Aug 16 10:09:49 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 16 Aug 1999 14:09:49 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au> <7p7aaj$sdl$1@cronkite.cc.uga.edu> <p6qhfm0nlzu.fsf@informatik.hu-berlin.de>
Message-ID: <7p963d$t9$1@cronkite.cc.uga.edu>

graham at sloth.math.uga.edu (Graham Matthews) writes:
: > : 	s = ""
: > : 	for  i in range(1,10): s = s + str(i)
: > : 
: > : Obviously i is an integer here, and s a string.
: > 
: > Is it? Surely that depends on what "range" and "+" are bound to at
: > runtime.
Martin von Loewis (loewis at informatik.hu-berlin.de) wrote:
: He said that in Viper, range is a keyword. How could you bind "+" to
: something else in the example?

You might have bound it in some code previous to this. If you are
planning to compile this code then you have to compile it so that it
will run in all contexts.

What I would like to know here is what is the type of range(1,10)?

graham
-- 

      England's not the mythical land of madame george and roses
        It's the home of police who kill black boys on mopeds



From herzog at online.de  Sun Aug  8 11:23:37 1999
From: herzog at online.de (Bernhard Herzog)
Date: 08 Aug 1999 17:23:37 +0200
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk> <slrn7qq4as.cc2.kas@titan.magnetic-ink.dk> <934111569snz@vision25.demon.co.uk>
Message-ID: <m3d7wymh3a.fsf@greebo.nodomain.de>

philh at vision25.demon.co.uk (Phil Hunt) writes:

> Yes. Does Python not allow this? What's the workaround? (Originally 
> my two modules were one, with hhh containing subclasses of bak's
> classes, and bak containing code to dynamically create instances
> of hhh classes at run time, using:
> 
>    newInstance = eval(nameOfClass + '()')
> 
> but I wanted the code split up into two modules as it is logically 
> separate.)

A good way to implement this IMO, is to have a registry of classes in
bak.py like this:

# bak.py
class BaseClass:
	# whatever
	pass

registry = {}

def instantiate(classname, *args, **kw):
	return apply(registry[classname], args, kw)


and in hhh.py:

# hhh.py
import bak

class SubClass(bak.BaseClass):
	# whatever
	pass

bak.registry['SubClass'] = SubClass


Code that has to create instances of SubClass would then call e.g.
bak.instantiate('SubClass')

This way, bak doesn't have to know anything about hhh or any other
module that implements sub-classes.



-- 
Bernhard Herzog	  | Sketch, a drawing program for Unix
herzog at online.de  | http://www.online.de/home/sketch/



From jeremy at cnri.reston.va.us  Thu Aug 19 14:20:03 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 19 Aug 1999 14:20:03 -0400 (EDT)
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc4352@cs.colorado.edu>
References: <199908191634.KAA01333@jhereg.perl.com>
	<37BC75F0.B482DA6D@starmedia.net>
	<37bc4352@cs.colorado.edu>
Message-ID: <14268.19155.505284.821500@bitdiddle.cnri.reston.va.us>

>>>>> "TC" == Tom Christiansen <tchrist at mox.perl.com> writes:

  TC>      [courtesy cc of this posting mailed to cited author] In
  TC> comp.lang.python, drew_csillag at geocities.com writes: :There are
  TC> the python manuals which are excellent, available at:
  TC> :http://www.python.org/doc/ (linked to from the homepage)
  TC> :available in your choice of HTML, PostSript, PDF, Info, LaTeX
  TC> and :windows help.

  TC> The fact that they did icky windows help and not man pages says
  TC> something significant.

Yup.  It says that Windows help is useful for more people than icky
manpages.  

(And who are these mysterious "they" people who did this?)

Jeremy



From mlh at idt.ntnu.no  Fri Aug 20 16:17:33 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 20 Aug 1999 22:17:33 +0200
Subject: making python man pages looks hard
References: <1103_935134343@shoal>
Message-ID: <y0j3dxe436q.fsf@vier.idi.ntnu.no>

connolly at w3.org (Dan Connolly) writes:

> 
> The source format is LaTeX. Converting from LaTeX is hard/messy:

Well - the docs aren't in standard LaTeX, are they? According to
www.python.org/doc/doc:

  "With almost no basic TeX or LaTeX markup in use, [...] the markup
  syntax is about the only evidence of LaTeX in the actual document
  sources."

Since the format seems to be clearly and simply defined, it shouldn't
be *that* difficult... Anyway - a transition to XML seems to be on its
way, which will make all this trivial. (Actually, it wouldn't be that
hard to go from the HTML versions either, I guess, with a little
imagination. My point is - there is no reason to be spooked by the
complexity of LaTeX here...)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From wtanksle at dolphin.openprojects.net  Sat Aug 21 15:16:11 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sat, 21 Aug 1999 19:16:11 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au> <7p7h3p$jto$1@newshost.accu.uu.nl> <slrn7rekon.r7j.wtanksle@dolphin.openprojects.net> <37bdda51.800579923@news.triode.net.au>
Message-ID: <slrn7rtunq.14r.wtanksle@dolphin.openprojects.net>

On Fri, 20 Aug 1999 22:47:42 GMT, John (Max) Skaller wrote:
>On Sun, 15 Aug 1999 23:54:00 GMT, wtanksle at dolphin.openprojects.net (William Tanksley) wrote:

>>Best of all, Pop-11 is finally free.

>	URL?

Er...  I don't know.  www.google.com...  Okay, here's an intro and
overview of some of the applications:
http://www.cogs.susx.ac.uk/users/adrianh/poplog.html; it neglects to
mention the recent open source release (the company shifted to other
things, so the license is XFree-like rather than GPL).
    ftp://ftp.cs.bham.ac.uk/pub/dist/poplog/freepoplog.html (took me
forever to find it) has the full info on the latest free versions.

The announcement post is at deja.com; search for "poplog GPL".

>John Max Skaller                ph:61-2-96600850              

-- 
-William "Billy" Tanksley



From fdrake at acm.org  Thu Aug  5 16:28:17 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 5 Aug 1999 20:28:17 GMT
Subject: Open3 / os.waitpid() bug
In-Reply-To: <7ocoho$pvp$1@nnrp1.deja.com>
References: <7ocoho$pvp$1@nnrp1.deja.com>
Message-ID: <14249.62433.659249.110770@weyr.cnri.reston.va.us>

sragsdale at my-deja.com writes:
 > In both cases, we didn't get the actual return value: we got something
 > much bigger.  I ended up discovering that for return value x, Python
 > thinks the child process returned 256*x.

  That's not *just* the return code; the Unix return code is that
number >> 8 (as you note); that number & 0xff tells HOW the process
exited (by a signal, calling exit(), other).  This is discussed in the
library reference somewhere, but I don't recall exactly where.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From m.faassen at vet.uu.nl  Wed Aug 11 04:21:14 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 11 Aug 1999 08:21:14 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au>
Message-ID: <7orbpq$i4s$1@newshost.accu.uu.nl>

John (Max) Skaller <skaller at maxtal.com.au> wrote:
> This is a request for comments on Viper, Yet Another implementation
> of Python. I'm building Viperi, the interpreter,  and I'm interested in
> comments on three issues:

> 	1) compatability with Python
> 	2) Extensions
> 	3) Optimisation

> Viperi is being built as a prelude to implementation of an optimising
> compiler, Viperc. Unlike previous attempts at a compiler, Viperc will do
> type inference, inlining, and special case analysis,
> in an attempt to optimise generated code.

It sounds like my (vaporware) Swallow idea. The difference with Viper
is that Viper apparently is going to do type inference, while Swallow
would (at least in the first stages) make do with nonintrusive type
declarations, somewhat like this:

__types__ = { 
  foo : functionType(args={a: intType, b: intType}, returns=listType(intType)),
  bar : functionType(args={}, returns=None)
}

def foo(a, b):
   return [a, b]

def bar():
   print "hello world"

But I forget the exact details right now. :) I did get a very simple Python
program going that could figure out if two complex types in the declaration
were the same. The idea is to go with a generic type approach somewhat
similar to C++ templates (but easier :). For instance, you'd say:

listType(intType) and Swallow would emit C code for a list of integers.

The idea is also to specify the types of local variables.

In the end, the only types that would need inferring with such a system
would be the type of empty lists and dictionaries:

__types__ = { foo : functionType(..., locals={ bar: listType(intType }) }

a = [] + [1, 2]

The first [] would have no type yet, but it would be fairly straightforward
to figure it out from the expression (I hope!)
 
> The interpreter aims to be compatible with 'pure python':
> python not using external C modules, and not fiddling
> implementation details too much. Some of the fiddles
> will work, and some will not. 

This is a good idea, as pure Python will often still be the development
language (while Swallow/Viper would only be invoked to compile and
optimize things later on, at least in the early stages).

> Compatibility:
> 	1) exceptions: both string and class exceptions are supported,
> 	the class exceptions are client defined in exceptions.py

> 	2) rebinding module variables after importing is not permitted
> 	3) None is a keyword.
> 	4) range is a keyword

Why this? Perhaps instead you could make 'range' and some other builtin
functions 'special functions'. The idea is that like module variables
(and some other set of things perhaps) they cannot be rebinded to something
else, so that the optimizer can do its stuff.

> 	5) What restrictions can you live with?

> Extensions:
> 	1) full slicing semantics
> 	2) optional values in dictionaries (Defaults to None)
> 	3) 'in' applies to dictionaries 'as if a sequence of keys were used'
> 	4) lexically scoped functions (including lambdas)
> 	5) optionally typed function parameters

> 	6) what do you want?

I wouldn't aim for *any* extensions to start with. In order for such a
project to succeed you want Viper code to run in Python and vice versa, 
unless there's serious messing with internals. At least make it possible
to turn these extensions *off* so they'll be errors.

[snip optimisation discussion]

Hm, I wish you luck, especially with the type inference...

Does viperc compile Python to C code? If so, try to minimize the amount of
C code that's emitted, if at all possible.

> 	Generally, compiler optimisations
> will go hand in hand with restrictions -- to optimise
> something, information is required, and that will
> require giving up some dynamism.

So you could do with Swallow's "type system" (big word for nothing there),
right? 

> 	It is difficult to optimise individual
> modules. It is a different story to optimise
> a whole program, where _every_ call
> to a function can be traced. Of course,
> this may involve restricting what 'exec' can do,
> and it may involve other restrictions
> (such as not  changing any module bindings
> after loading).

Hm, but the possibility to optimise individual modules is what attracts
many people to this idea. This way, people can stay away from writing
C modules and write them in Python instead. The Swallow cycle I had in
mind was something like this:

  * Write Python module

  * Need to make it faster (at least some functions)

  * Supply full type information for any local variables, arguments, etc.
    Anything *called* by the functions should also be Swallowable (or already
    supported by Swallow).

  * Swallow now optimises these functions by translating them to C, and
    you can use them in Python (just like C extension modules).

For classes you'd need some extra support (and more restrictions) but that
was phase 2. My advice is to start as small and unambitious as possible
for such an ambitious project. :)

Regards,

Martijn




From paul.black at nist.gov  Wed Aug 25 12:24:29 1999
From: paul.black at nist.gov (Paul E. Black)
Date: 25 Aug 1999 12:24:29 -0400
Subject: word1.word2 notation
References: <oxgbtbxejea.fsf@panix7.panix.com>
Message-ID: <2llnazj0aq.fsf@whitestone.ncsl.nist.gov>

Yeoh Yiu <squid at panix.com> writes:

> I've noticed in various languages, there is a construction
> of expressions with a dot in the middle.
> 
> 	int foo.bar = 42;   // doesn't happen this way
> 
> What is the name of this technique and the name of methods 
> (or variables ?) so constructed ?

The dot construction is merely syntax, and, as you have observed, is
used in different languages.  The syntax indicates different semantics
in different languages and therefore has different names.

-paul-
-- 
Paul E. Black (p.black at acm.org)



From lars at pixar.com  Fri Aug 20 19:22:18 1999
From: lars at pixar.com (Lars Damerow)
Date: Fri, 20 Aug 1999 16:22:18 -0700
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc4111@cs.colorado.edu>
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu>
Message-ID: <Pine.SGI.4.10.9908201606400.525821-100000@grendel>

On 19 Aug 1999, Tom Christiansen wrote:

> No, that's not it.  Perl's hex takes a hex number and returns 
> a decimal one; same with oct.  Python's is the reverse.  Very weird.
> 
>     echo 'print oct(12)' | python
>     14
> 
>     echo 'print oct(12)' | perl -l
>     10
> 
> See what I mean?

I don't consider that weird at all.  Take casting, for example, in any number
of languages.  If, in C++, I say int(var), that takes var and casts it to be an
integer.  I might say long(num) in Python, which says "Turn num into a long."

In that case, doesn't it make sense that hex(12) would take the decimal number
12 and return '0xc'?  That oct(153) returns '0231'?  Functions and methods are
typically named for the action they perform, not the data they're accepting.

-lars

___________________________________________________________
lars damerow
technomonkey
pixar animation studios
lars at pixar.com

"Okay, there's the house.  I'm not stopping the car, so you'll have to
jump out."





From mlh at idt.ntnu.no  Tue Aug 31 19:10:16 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 01 Sep 1999 01:10:16 +0200
Subject: Fuzzy string matching?
References: <y0j7lmiuyo2.fsf@vier.idi.ntnu.no> <19990827231353095.AAA161.428@max41051.izone.net.au> <y0jso53udoo.fsf@vier.idi.ntnu.no> <slrn7sgbut.lbd.thantos@brimstone.mecha>
Message-ID: <y0jzoz74kdj.fsf@vier.idi.ntnu.no>

thantos at brimstone.mecha (Alexander Williams) writes:

> On 28 Aug 1999 17:28:55 +0200, Magnus L. Hetland <mlh at idt.ntnu.no> wrote:
> >Could perhaps be a bit more readable, though... (Maybe I'll work on
> >that...)
> 
> Hey, um, while we're talking readability, this is where I grind my own
> personal axe about variable names consisting of a single, solitary
> character, isn't it?  :)
> 

Hey!

Take a look at my latest version...

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From wtanksle at dolphin.openprojects.net  Mon Aug 16 16:50:36 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Mon, 16 Aug 1999 20:50:36 GMT
Subject: Python CGI Scripts
References: <Pine.LNX.4.04.9908162019270.3026-100000@natal.stamp-coin.com>
Message-ID: <slrn7rgucs.udq.wtanksle@dolphin.openprojects.net>

On Mon, 16 Aug 1999 20:24:42 +0000 (UTC), Jonathon wrote:
>On Mon, 16 Aug 1999, Rudi Angela wrote:

>>Yep, have a look at Zope, a python based opensource cgi solution:

>	From my POV, Zope has two major problems.

Only two :)?

>	#1:	It expects to run as a server.  This is not a good
>		thing, from the POV of most IPPs and ISPs.

I'm confused -- since its earliest release it's run under Apache.  What
could you mean "run as a server"?  Were you expecting it to be a client?

>	#2:	The documentation is somewhat lacking.
>		<< This one is easy to fix. >>	

You nailed that one right on.  It's not easy, but it sure is important.

>	I suspect that the same would be true for various other
>	people as well.

I'm sorry, what?  What are you saying would be true for people?

>	A Python equivelent to the Perl CGI.pm module would be a
>	very usefull thing to have/do.

I'm pretty sure it's already there, even for people who don't use Bobo.
People have been using Python for CGI scripting for a while.

>	xan
>	jonathon

-- 
-William "Billy" Tanksley



From daniel.dittmar at sap-ag.de  Mon Aug 16 08:18:29 1999
From: daniel.dittmar at sap-ag.de (Daniel Dittmar)
Date: Mon, 16 Aug 1999 12:18:29 GMT
Subject: Extension types as exceptions (Wrapup)
References: <7our22$9fb$1@mailusr.wdf.sap-ag.de> <mYDs3.1626$EG4.303931@news.shore.net>
Message-ID: <7p8vj5$i0s$1@mailusr.wdf.sap-ag.de>

Thanks for the help,

for those with similar problems, code to create the exception type
follows:

static PyObject *
createExceptionKind (
    const char * exceptionName,
    const char * exceptionCode)
{
    PyObject * result;
    PyObject * runResult = NULL;
    PyObject * globals = NULL;
    PyObject * locals = NULL;

    if (exceptionCode != NULL) {
        globals = PyDict_New ();
        locals = PyDict_New ();
        runResult = PyRun_String ((char *) exceptionCode,\
Py_file_input,
            globals, locals);
    }
    result = PyErr_NewException ((char *) exceptionName, \
NULL,locals);
    Py_XDECREF (locals);
    Py_XDECREF (globals);
    Py_XDECREF (runResult);
    return result;
}

Parameter exceptionName is something of the form "mymod.ExceptionName"
Parameter exceptionCode is something like
"def __str__ (self):\n    return 'ExceptionName: %d %s' % (self.code,
self.msg\n"

Error handling is a bit weak as I call this from module
initialization, which does it's own checking

Code to create a new exception:

static void
raiseError (
    int code,
    const char * msg)
{
    PyObject * exception = PyInstance_New (ErrorType, NULL, NULL);

    if (exception == NULL) {
        return;
    }
    PyObject_SetAttrString(exception,"code", PyInt_FromLong  (code));
    PyObject_SetAttrString(exception,"msg", PyString_FromString(msg));
    PyErr_SetObject (ErrorType, exception);
}

ErrorType is a c module static which contains the result of
createExceptionKind





From kernr at mail.ncifcrf.gov  Fri Aug 20 17:12:55 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Fri, 20 Aug 1999 21:12:55 GMT
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <Pine.LNX.4.10.9908200706500.15847-100000@starship.python.net>
Message-ID: <37bdc47f.22030506@news.erols.com>

On Fri, 20 Aug 1999 07:08:16 -0400 (EDT), Moshe Zadka
<moshez at server.python.net> wrote:

>On Fri, 20 Aug 1999, Greg Ewing wrote:
>
>> Martijn Faassen wrote:
>> > 
>> > Funny, I read that 'the way g-d indented'. :)
>> 
>> Hey, what a great signature that would make!
>> 
>> Python: Programming the way g-d indented.
>
>Make that:
>Python: Programming the way g--d- indented.

Why does this remind me of Douglas Hofstadter's _G?del, Escher, Bach_?
:-)

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From xah at best.com  Sun Aug 22 00:00:54 1999
From: xah at best.com (Xah)
Date: Sat, 21 Aug 1999 21:00:54 -0700
Subject: Why use Python when we've got Python?
References: <37b4ba01@cs.colorado.edu> <37B50C9D.868F0F02@inka.de>
	 <37b56fcf@cs.colorado.edu>
Message-ID: <37bf75f7$0$207@nntp1.ba.best.com>


----------
>From: Tom Christiansen <tchrist at mox.perl.com>
>Newsgroups: comp.lang.perl.misc,comp.lang.python
>Subject: Re: Why use Python when we've got Perl?
>Date: 1999, Aug 14, , 7:31 AM

>
> Certainly Perl's hooks into Unix are stronger than are its hooks for
> proprietary systems from Microsoft, Apple, or the erstwhile DEC, which
> even if existent, tend to manifest themselves as add-on modules rather
> than as original core functionality.  This of course makes perfect
> historical sense, given the natal environment and concomitant problem
> domain.
>
> Nevertheless, "Perl: The Programmer's Companion" by Nigel Chapman
> (whose ambiguously worded title actually posits Perl as a companion
> of programmers) does a commendable job of providing a solid technical
> foundation for the language with nary a whit of system-specific arcana.
> This small and inexpensive volume (in which I aver no personal interest)
> stands head and shoulders above all others as the best introductory Perl
> book, especially for real programmers such as yourself.  Script kiddies,
> tape operators, and all such ilk occasionally merit other recommendations,
> some of which occasionally include rather than a book title something
> on the order of, "Hey kid, hurry up with those french fries!" :-)
>
> --tom

You realize that book does the Perl community a de-service? It doesn't talk
about unix, the author uses MacPerl, it's written with clarity, real humor,
and exceptionally well organized. The worst is that it's written by an
computer scientist. <shudder>

No, I rather prefer YOUR opuses. Besides, yours are so popular and
voluminous that I can take it with me when I visit john and use it to wipe
my ass with spare.

I'm thinking of writing a comprehensive, free, quality Perl on-line manual
for the GNU project. Would you help out? Y'know, without your interjection,
I'll probably end up writing something like that Chapman fellow's.

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 "Perl my ass -- there are more than one way to express love too."
      --Larry Wall, from antimatter universe.



From andy at robanal.demon.co.uk  Tue Aug 10 15:54:52 1999
From: andy at robanal.demon.co.uk (Andy Robinson)
Date: Tue, 10 Aug 1999 19:54:52 GMT
Subject: International characters and Python's string functions on Linux
References: <MPG.1213b1f2ede568ee989681@news.mop.no> <37aa0196.1944357548@news.triode.net.au>
Message-ID: <37b18350.5166148@news.demon.co.uk>

>
>	http://www.triode.net.au/~skaller/unicode/index.html
>
>for documented sources. 
>
>	In Python today, you can support
>internationalisation using the UTF-8 encoding, as my 
>literate programming tool interscript does. UTF-8 is,
>in my opinion, the best option in Python today, since it
>is ASCII compatible, and will work with 8 bit strings as 
>Python today already has.

Internationalisation is not just about which encoding you use, it is
about having the libraries in your app to convert between them as
needed.  

I use Python to build live gateways between Sybase servers holding
7000+ Japanese names and addresses in UTF8, the same data on AS400s in
IBM's own undocumented encoding, and Shift-JIS on Windows.  We have to
handle all three, as those are the systems we have to interface to.
We also have to encode things the way our printers like for printing.


If you have a saner problem to deal with (European languages and one
OS), the key issues are still what encodings do you get data from,
what encodings do you have to save it to, and do you have to print it?
The choice of encoding is never made in isolation.

John Skaller's conversion utilities are a great start - I had
something uncannily similar but it was written for clients and I
cannot release it.  What Python really needs to handle
internationalisation well is not an even better wide string type, but
a standard library to convert all those other encodings to and from
UCS2/UTF8.  Java does this out of the box.


Regards,

Andy




From tim_one at email.msn.com  Fri Aug 20 01:34:57 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 20 Aug 1999 01:34:57 -0400
Subject: slice indices (RE: More random python observations from a perl programmer)
In-Reply-To: <37BC2D28.46D1FA8E@easics.be>
Message-ID: <001201beeacd$bd088500$712d2399@tim>

[tchrist]
> > GOTCHA: (medium)
> >     All ranges are up to but *not including* that point.  So range(3)
> >     is the list [0,1,2].  This is also true in slices, which is the
> >     real gotcha part.  A slide t[2:5] does not include t[5] in it.

[Jan Decaluwe]
> I always found this very strange. However, in practical use it turns
> out to be often quite natural, for example:
>    a[:5] returns 5 elements.

More generally, seq[i:j] contains j-i elements (assuming 0 <= i <= j).

> Exactly the same approach is used by the containers in C++'s STL
> library, so I guess there must be some good theoretical arguments
> also.

It's not theory, it's practical <wink>:  learn to view indices as pointing
*between* elements, and it all makes perfect sense at once.  Where does

    x.insert(5, thing)

insert thing?  If you view indices as pointing *at* elements, it's unclear
whether "thing" ends up to the left or the right of x[5].  But view 5 as
pointing *between* adjacent elements, and it's obvious.

Ditto for sequence[i:j], and so on.  Viewing that as "the slice beginning at
i and ending before j" is hopelessly convoluted.

True story:  Fortran was my first high-level language.  When I learned C, I
looked at e.g.

int c[10];

and wondered why on earth C declared arrays with a number one higher than
the largest legit index!  It was almost a year before it dawned on me that
it was meant to be viewed as declaring the total number of elements.

Python's flavor of slice notation will look just as obvious and natural when
it "clicks" for you.

otoh-i-still-don't-know-where-to-put-the-damned-"const"s-ly y'rs  - tim





From shecter at darmstadt.gmd.de  Thu Aug 19 13:32:24 1999
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Thu, 19 Aug 1999 19:32:24 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
		<slrn7ro8n1.h6v.jblaine@shell2.shore.net>
		<37bc29d8@cs.colorado.edu> <14268.15201.81468.481452@bitdiddle.cnri.reston.va.us>
Message-ID: <37BC3FA8.9937211D@darmstadt.gmd.de>

Jeremy Hylton wrote:

> 
> You can do almost exactly this in Python using the update method on
> dictionaries.
> 
>     x = {1:2, 3:4}
>     y = {10:20, 30:40}
>     z = {}
>     z.update(x)
>     z.update(y)
>     print z
>     {30: 40, 3: 4, 10: 20, 1: 2}
> 
> As you noted with the [].sort operation, you need to explicitly create
> the new object that call some methods on it.  In my book it is at
> worst a minor annoyance.

In my book it's the right way to do things.  My book, by the way, is
"Object Oriented Software Construction", by Bertrand Meyer.  He makes
the great point for developing classes that work in a consistent
style:

1) You instantiate an object
2) You set some options
3) You tell it to do its thing

This pays off big in clarity of code.  Related to this are his other
commandments, such as banning functions with side effects.

Perl's version:

%z = (%x, %y);

is ambiguous.  It looks more like it creates a nested structure than a
flat one.

- Robb



From skip at mojam.com  Thu Aug 12 14:40:47 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 12 Aug 1999 13:40:47 -0500 (CDT)
Subject: Copying Huge Files ? (Newbie!)
In-Reply-To: <37b30e52.43353218@news.usenetserver.com>
References: <37b30e52.43353218@news.usenetserver.com>
Message-ID: <14259.5159.987311.942071@dolphin.mojam.com>

    Matthias> But while reading the manual I cannot find something like
    Matthias> os.copyfile or os.movefile.

    Matthias> Are there some functions except open/read/write that can do
    Matthias> the copy for me ???

Matthias,

You'll have to open the input and output file, then read and write chunks
yourself, for example (untested!):

    def copy(fromfile, tofile, chunksize=8192):
	f = open(fromfile, "rb")
	t = open(tofile, "wb")
	data = f.read(chunksize)
	while data:
	    t.write(data)
	    data = f.read(chunksize)
	f.close()
	t.close()

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From skaller at maxtal.com.au  Fri Aug 20 17:40:14 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Fri, 20 Aug 1999 21:40:14 GMT
Subject: Extending Python: exception handling
Message-ID: <37bdcb30.796706964@news.triode.net.au>

 I would like to know how YOU are using
exception handling OTHER than for error management.

Reason: there is no such thing as an invalid python program.
Therefore, python cannot be extended, not even by Guido :-)

[this post is valid python -- it throws a SyntaxError, which is
well defined behaviour]

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From jeremy at cnri.reston.va.us  Wed Aug 18 13:12:58 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Wed, 18 Aug 1999 13:12:58 -0400 (EDT)
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <14266.49377.51221.305982@dolphin.mojam.com>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<14265.47748.241386.291144@buffalo.fnal.gov>
	<wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
	<m3672dwojr.fsf@pcrm.win.tue.nl>
	<14266.49377.51221.305982@dolphin.mojam.com>
Message-ID: <14266.59802.56315.731454@bitdiddle.cnri.reston.va.us>

>>>>> "SM" == Skip Montanaro <skip at mojam.com> writes:

  Stephan> Furthermore, Skip used he module whrandom, whereas I used
  Stephan> random.  Any ideas in which respect these modules differ?

  SM> Well, if you look at random.randrange, you'll find it
  SM> exceedingly similar to whrandom.randrange:

  >>>> import whrandom, random random.randrange
  SM>     <method whrandom.randrange of whrandom instance at 814a370>
  >>>> whrandom.randrange
  SM>     <method whrandom.randrange of whrandom instance at 814a370>

The intended use of the standard library here is for users to import
random to get access to whrandom.  whrandom implements a specific
method for generating random numbers.  It could easily be replaced
with a different, improved module at some time in the future.  If user 
code imports random, it will automatically get the benefit of the new
implementation. 

Jeremy



From sholden at pgrad.cs.usyd.edu.au  Fri Aug 13 22:32:12 1999
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 14 Aug 1999 02:32:12 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ce93@cs.colorado.edu>
Message-ID: <slrn7r9la9.pu8.sholden@pgrad.cs.usyd.edu.au>

On 13 Aug 1999 20:04:03 -0700, John W. Stevens <jstevens at basho.fc.hp.com> wrote:
>> In comp.lang.perl.misc, 
>>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
>> :$b[ 2 ] = $c;
>> :
>> :> That's just fine in Perl.  It's not fine in Python, because Python
>> :> won't automatically grow an array.
>> :
>> :'Cause it doesn't have arrays (or, at least, not built in ones).
>> 
>> Gosh, that's a feature.  NOT.
>
>Perl doesn't have lists.  Python doesn't have built-in arrays.

You should learn some perl you now..

@array = (1,10,20,30);
$from_list = (1,10,20,30);
$from_array = @array;
print "$from_list\n$from_array\n";

Will output :
30
4

Perl has lists, if you know perl you would know this. If you program in perl 
and don't know this, then you must get very very confused at times.

>
>Python, however, has an array module as part of its standard library.
>
>I will assume that a list module is available for Perl.

No it is one of the built in bits... like hashes and arrays.

>
>I wasn't trying to compare features, I was simply pointing out
>that your comparison was Apples and Oranges, and therefore at
>least somewhat invalid.

Only because you have no idea what you are talking about.

-- 
Sam

Can you sum up plan 9 in layman's terms? It does everything Unix does
only less reliably.
	--Ken Thompson



From m.faassen at vet.uu.nl  Thu Aug 19 18:04:29 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Aug 1999 22:04:29 GMT
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk> <19990816205639474.AAA315.413@max41127.izone.net.au> <lDvYnDAz1Hu3EwB6@jessikat.demon.co.uk> <37BB2E1D.C0733F85@boeing.com> <7pgd1g$hj6$1@newshost.accu.uu.nl> <37BC7014.33B6FFF6@compaq.com>
Message-ID: <7phv1d$nvo$1@newshost.accu.uu.nl>

Greg Ewing <greg.ewing at compaq.com> wrote:
> Martijn Faassen wrote:
>> 
>> Funny, I read that 'the way g-d indented'. :)

> Hey, what a great signature that would make!

> Python: Programming the way g-d indented.

*laughs* 

Go and use it!

Regards,

Martijn




From tim_one at email.msn.com  Tue Aug 31 21:28:08 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 1 Sep 1999 01:28:08 GMT
Subject: Soundex implementation (was: RE: Fuzzy string matching?)
In-Reply-To: <19990831123108417.AAA352.267@max41072.izone.net.au>
References: <19990831123108417.AAA352.267@max41072.izone.net.au>
Message-ID: <000801bef419$3f5be7a0$31a0143f@tim>

[Tim, on Soundex]
>| I find Knuth's description of the algorithm ambiguous in several
>| respects,

[John Machin]
> I see no ambiguity here; certainly, it is very hard to see anything
> so ambiguous as to give you the leeway to treat W and H as different
> from Y! Please elaborate.

That part isn't ambiguous.  I don't know which person you thought of when I
wrote "Knuth", but the guy I'm thinking of makes a special case of W and H (but
not Y) in step 3:

   3. If two or more letters with the same code were adjacent in
      the original name (before step 1), or adjacent except for
      intervening h's and w's, omit all but the first.

If you can read that as saying W and H are the same as Y, the ambiguities are
even worse than I had suspected <wink>.

>| and there are many incompatible soundex algorithms "out
>| there".

> So Python should have yet another incompatible soundex algorithm?
> I didn't think me-too-ism was the Python way. My preference would
> be to implement what's in Knuth's book, subject to resolution of
> the ambiguity question.

I did, to the best of my understanding.  If your understanding differs, write
an implementation that matches yours, and we can claw each other to death over
which is the True Knuthian Algorithm <wink>.

>| soundex-is-a-pragmatic-hack-so-right-vs-wrong-is-
>| fuzzy-ly y'rs - tim

> I agree entirely but I don't think your WY hack improves the result --
> - please supply some counter-examples if you have any ---

As above, it was Knuth's "hack", but I don't think it's hard to see the
motivation:  vowels between similar consonants generally "break up" the latter
aurally, but in English "h" and "w" generally aren't voiced intraword while "y"
generally is (and Knuth is treating all of "why" as vowels in step 1).  An
example is his own Wachs -> W200.  Here c and s both map to 2, but the step-3
rule correctly predicts that the "h" between them "can't be heard".  OTOH,
Wacys is more likely to sound like Wacis or Wakoz:  an intraword vowel is
generally voiced in English, and then W220 is the proper code.

> but can have a detrimental effect with a few Welsh names; see below.

Like Uncle Don gives a hoot about Wales <wink>.

saw-a-hebrew-variant-of-soundex-once-that-slobbered-on-for-
    pages-ly y'rs  - tim





From sjmachin at lexicon.net  Fri Aug 27 19:18:47 1999
From: sjmachin at lexicon.net (John Machin)
Date: Fri, 27 Aug 1999 23:18:47 GMT
Subject: Fuzzy string matching?
In-Reply-To: <y0j7lmiuyo2.fsf@vier.idi.ntnu.no>
References: <y0j7lmiuyo2.fsf@vier.idi.ntnu.no>
Message-ID: <19990827231353095.AAA161.428@max41051.izone.net.au>

On 26 Aug 99, at 21:31, Magnus L. Hetland wrote:

> mlh at idt.ntnu.no (Magnus L. Hetland) writes:
> 
> > "Hans Nowak" <hnowak at cuci.nl> writes:
> > 
> > > ...Anyone know of such a beast in Python?
> [...]
> > Other than that - fuzzy matching is a non-trivial problem. It is
> > possible to use dynamic programming to make a quadratic time distance
> > function that you can use to compare words or strings - to see how
> > similar they are. (If you like, I can post a Python variant of it.)
> 
> I decided not to wait for your answer... :)
> 
> def distance(a,b):
> [[ SNIP ]]

> 
> (And if anyone wants an explanation as to how this distance function
> really works, I'll be glad to supply it... I've done lectures about it
> in an algorithm class - and will be doing so later this fall as well...)

Magnus,
One presumes that your first assignment for your students is to go to 
the library and/or think a little and come back with the well-known 
O(min(m,n)) instead of O(mn) space requirement optimisation to the 
algorithm you presented and therefore you would prefer not to have it 
posted to this list, just in case they read it. Oh and for what it's 
worth, although the time is still O(mn), it decreases --- I have yet to 
see a language where foo[i,j] takes no more time than bar[j]. Python, 
with foo being a dictionary and bar being a list, is definitely not 
such a language.
Cheers,
John




From chad at vision.arc.nasa.gov  Mon Aug  9 00:53:23 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Mon, 09 Aug 1999 04:53:23 +0000
Subject: RPM for Python 1.5.2 ?
References: <slrn7qrn3u.7fj.dyoung@baltazar.baltazar.org>
Message-ID: <37AE5EC3.459676D5@vision.arc.nasa.gov>

David Young wrote:

> Has anyone built an RPM for Python 1.5.2 ?

Someone else pointed you to Oliver's RPMs, for RH 5.2.  Oliver started a list
to discuss packaging issues,
and he (and I and others) are working on getting a version of 1.5.2 that
works on RH 6.0, Mandrake 6, and
others.

To see the list archives, go to:
http://starship.python.net/mailman/listinfo/python4linux .  I hope to get
some time this week to fix Redhat's pythonlib package so that it works
properly with 1.5.2.  For an example
of why it DOESN'T work now, check out this line from listbox.py in the
/usr/lib/rhs/python directory:

    def __init__(self, Master = None, fieldList):

Other examples of using non-default arguments after defaults exist, and
should be easy to fix.  I don't
understand how the redhat 1.5.1 python package doesn't break (I don't have it
installed, and I haven't
had a chance to check into it...)

Chad





From noreply at domain.com  Mon Aug  2 17:42:21 1999
From: noreply at domain.com (Brian)
Date: Mon, 2 Aug 1999 14:42:21 -0700
Subject: JPython feedback
Message-ID: <8mop3.20$o3.5158@news.premier1.net>

Hello!
Is anyone here actively using JPython?  Any positive/negative feedback of it
would be appreciated.  Have you been able to develop any large applications
with it?  Any "gotcha's"?
Thanks!






From cgw at fnal.gov  Thu Aug 19 11:58:31 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Thu, 19 Aug 1999 15:58:31 GMT
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <000701bee917$d8161cc0$eaa22299@tim>
References: <000701bee917$d8161cc0$eaa22299@tim>
Message-ID: <14268.10663.137315.893101@buffalo.fnal.gov>

Tim Peters:
 > Me:
 > > ...
 > > Another, more straightforward way to test sampling strategies is just
 > > to keep track of how many times each permutation comes up.  However,
 > > since N! gets so large, it takes too long to build up decent
 > > statistics this way.
 > 
 > I'll save you a near-infinity of work, then <wink>:  Python's Wichmann-Hill
 > uniform generator has a period of around 1e12 to 1e13. 

Thanks for the labor-savings.  The point I was getting at however was
that even with a perfect random number generator, if you're not
careful, you can write a function to turn those random numbers into
permutations which does not evenly distribute across the space of
permutations - for example favoring even permutations over odd
permutations, etc.  The goodness of the random-number generator is
another issue...

I really should go off and read more Knuth; as Dan Schmidt pointed out
all of this is discussed there; rather than devising clever
statistical tests, it's far better to *prove* from first principles
that the algorithm is correct.








 Since the point at
 > which you begin in the period completely determines the permutation that
 > will be generated, you can't get more than 1e13 distinct permutations out of
 > it.  So a sequence of length 16 is the longest for which it could even
 > conceivably produce all permutations; if you use it to, e.g., shuffle a deck
 > of cards, the number of permutations any algorithm based on WH can get back
 > is a vanishingly small fraction of the possibilities (since 52! ~= 1e66; we
 > can only get back at most 1 of each 1e66/1e13 = 1e53 possibilities).
 > 
 > So, in a real sense, even the best possible algorithm based on
 > random.random() is terrible, incapable of generating "almost all" of the
 > space for a measly deck of cards.
 > 
 > It's surprising that so few people get disturbed by this; but it's even more
 > surprising that some people do <wink>.
 > 
 > the-deck-is-always-rigged-in-subtle-ways-ly y'rs  - tim
 > 
 > 
 > 
 > 



From news at dorb.com  Tue Aug 31 13:25:09 1999
From: news at dorb.com (Darrell)
Date: Tue, 31 Aug 1999 13:25:09 -0400
Subject: Python speed and `pcre'
References: <oqzoz89hrb.fsf@titan.progiciels-bpi.ca>
Message-ID: <HdUy3.3199$v6.166809@typhoon.nyroc.rr.com>

>From re.py
_MAXCACHE = 20

Definitely do the profile before assuming where the expense is. I've found a
common mistake is splitting and combining strings. Use string.join(), don't
do this str=str+str1 in a loop. Strings are immutable so you end up
allocating memory and coping every time. Very expensive.

--
--Darrell
Fran?ois Pinard <pinard at iro.umontreal.ca> wrote in message
news:oqzoz89hrb.fsf at titan.progiciels-bpi.ca...
> Hi, people.  Still learning and experimenting with Python. :-)
>
> After having translated some code (not big, but not small) from Perl to
> Python, I discover it runs ten times slower.  I did not learn to profile
> yet, and I am not sure it would help, after having read that profiling
> gives call counts, but no elapsed time.  Is that right?
>
> My intuition tells me that the amount of regular expression matching might
> provide an explanation.  Here is my set of _hypotheses_ (I'm not sure):
>
> * Perl keeps compiled all /REGEXP/ not using string interpolation,
> * Python cache for compiled REGEXP is less virtuous than I imagined.
>
> I would be tempted to try using, in Python, a trick like `gettext' does
> to avoid retranslating in already translated message.  That is, for _each_
> textual `gettext' call, the compiler #define extends some code statically
> allocating of a local variable used to cache the result or the call,
> and use this cached result afterwards.
>
> My problem, and I hope you will have some good advice to give for it:-),
> is that I do not see how to this legibly.  I could regroup a lot of
> `re.compile' calls and use the corresponding variable afterwards, but I
> would much rather keep the REGEXP expressions textual near the place they
> are needed, and have some:
>
>         global gensymed-local
>         if gensymed-local is None:
>     gensymed-local = re.compile(REGEXP)
>         gensymed-local.sub(...)
>
> This would have the advantage of keeping the REGEXP where it is meaningful
> in the code.  Yet, creating and maintaining all those `gensymed-local'
> variables by hand might be fairly tedious, not to say some code to ensure
> they are all None.  (But maybe `global' already guarantees that?  If yes,
> may I count on it, that is, does Python promises it is unlikely to
change?)
>
> Of course, seeing the above clutter, the idea a macro-generator is surely
> tempting, but this would be fairly heavy resorting to external mechanics
> for something which should be straightforward.  If Python included a
> macro-generator, it would be a nice thing...
>
> Is there something fundamental I am missing, and some just simpler code
> that would have the desired effect?  Or else, is there another approach
> that could give me more speed without giving too much on legibility?
> I'm surely ready to accept some slowdown going from Perl to Python, but
> a slowdown factor or 10 is a bit much, in my opinion.
>
> --
> Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard
>
>
>





From fredrik at pythonware.com  Sun Aug 22 12:44:20 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 22 Aug 1999 16:44:20 GMT
Subject: can some one help a newbie?
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <7pkke6$lut@dfw-ixnews6.ix.netcom.com> <935227166snz@vision25.demon.co.uk> <7pmv4j$gtt@dfw-ixnews14.ix.netcom.com> <935273107snz@vision25.demon.co.uk>
Message-ID: <00fe01beecbd$95da5340$f29b12c2@secret.pythonware.com>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
> The idea that a company would employ someone, who can't
> really code, as a sysadmin is IMO frightening. Unless they
> don't care whether their computer system works, of course.

...

a friend working for a big consultancy firm claims
they're handing out the following questionnaire to
anyone who happens to walk by outside their
office:

1. have you ever seen a computer: yes/no

2. if you answered yes, have you seen a
computer in real life (and not just on tv,
in a magazine, etc): yes/no

one yes and you get a job, two yes and you're
hired as an expert consultant.

...

also see:
http://www.python.org/doc/essays/cp4e.html

</F>




From jim at interet.com  Fri Aug 20 09:05:27 1999
From: jim at interet.com (James C. Ahlstrom)
Date: Fri, 20 Aug 1999 09:05:27 -0400
Subject: How to sell Python to your boss?
References: <ka8yaf7d84t.fsf@delightful.cs.hut.fi>
Message-ID: <37BD5297.9AF6B5DA@interet.com>

Petri Mikael Kuittinen wrote:

> rudimentary, but easy-to-use, GUI for the multi-player game. My boss
> otherwise agreed with my idea ("it would be really useful..."), but he
> doesn't want me to do the prototype implementation using Python. He
> said I must use a program language that all the other programmers in
> our R&D department understand. Well, that leaves only one choice: C

Another thought.  Agree to write it in C, but with a useful
free library add-on utility which is also written in C 
called "Python".

Jim Ahlstrom



From chad at vision.arc.nasa.gov  Thu Aug 12 11:43:34 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Thu, 12 Aug 1999 15:43:34 +0000
Subject: Problem when using FileDialog from PMW Dialog.  Quick fixes?
References: <37B21CE1.488F45FA@vision.arc.nasa.gov> <37b2cb01.501756387@news1.on.sympatico.ca>
Message-ID: <37B2EBA6.5FDD6FD7@vision.arc.nasa.gov>

"Robert J. Roy" wrote:

> If this code snippet is accurate

It isn't, in my haste to make an example I accidentally reused the dialog
variable instead of making a separate one for the nested dialog.  I noticed
it a while after I posted, when I was creating a related demo for Greg
McFarlane.  However, due to chance, it doesn't affect the demonstration,
because the binding and callbacks have been registered by the time dialog
gets remapped.  At least on my system (Linux), the same result occurs
with either my original code, or your revised code (ie. the Pmw.Dialog
window does not respond to button selections anymore)  I did some
more tests, and found this behavior to be somewhat common.  (Greg
pointed me to PmwFileDialog in the Pmw contribs, and I was able to
get similar behavior by triggering modal dialogs to popup during its
activation.)


> You might want to look at using tkFileDialog instead. It brings up the
> standard file dialogs, presenting a more familiar interface to Windows
> users.

Ah yes, I have used it in the past, and when I went looking for it, I forgot
that it isn't in the Tkinter module.  Having the native look would be
useful...  Let me test it out [5-10 min later]  Yes, it works fine nesting
with a Pmw widget.  I'd forgotten how ugly that requestor looks on Unix,
though.  :)  Cheers, I'll keep it in as an option for now.

Thanks for the help.

Chad
chad at vision.arc.nasa.gov





From chad at vision.arc.nasa.gov  Wed Aug 25 17:40:12 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Wed, 25 Aug 1999 21:40:12 +0000
Subject: PIL: problem with im.show()
References: <7q1f58$r7v$1@tribune.mayo.edu>
Message-ID: <37C462BC.A159FB27@vision.arc.nasa.gov>

Tom Kinter wrote:

> system configuration
>   Sun Solaris-2.6
>   python-1.5.2
>   PIL-1.0
>
>
>   File "/usr/local/lib/python1.5/site-packages/PIL/Image.py", line 311, in _dump
>     file = file + "." + format
> TypeError: illegal argument type for built-in operation
> >>>
>

I had the same problem on a Linux box when I was creating an SRPM for Imaging 1.0.
I created a patch to fix it, and sent that to F. Lundh (no response yet).

In any case, the following patch fixed it for me:

--- PIL/Image.py.orig   Mon Aug  2 15:39:56 1999
+++ PIL/Image.py        Sun Aug 15 14:30:28 1999
@@ -961,7 +961,7 @@
        if not command:
            command = "start"
     else:
-       format = None
+       format = "PPM"
        if not command:
            command = "xv"
            if title:

In other words, on Unix, the default format string defaults to None.  Change it
to a valid extension string (if "PPM" doesn't work for your viewer, try "TIF"
or "BMP" or "JPG" or whatever)

Chad
chad at vision.arc.nasa.gov





From bernhard at alpha1.csd.uwm.edu  Sat Aug 14 04:12:53 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 14 Aug 1999 08:12:53 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov>
Message-ID: <slrn7ra985.s26.bernhard@alpha1.csd.uwm.edu>

On 13 Aug 1999 19:41:18 GMT, Drew McDowell <drew.mcdowell at msfc.nasa.gov> wrote:
>I looked in the FAQ and couldn't find anything on the Compound
>Assignment Operators.
>(+=, *=, -=, etc..)
Yeah, maybe it should be a FAQ.
>Why dosen't Python support them?  
Don't really know, I guess the authors consider them unnecessary
syntactic sugar.
>Is there a easy way to add them?
Make you editor replace += with the proper thing.
Browse newgroup archive (dejanews) to find some example for vim...
	Bernhard


-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.    (www.uwm.edu/~bernhard)
Association for a Free Informational Infrastructure              (ffii.org)



From donn at u.washington.edu  Fri Aug 27 13:12:12 1999
From: donn at u.washington.edu (Donn Cave)
Date: 27 Aug 1999 17:12:12 GMT
Subject: how to get output from system call into data object
References: <863dx5gmrr.fsf@tocol.chc.us.ac.com> (none)
Message-ID: <7q6gtc$1038$1@nntp6.u.washington.edu>

Quoth stephen at tocol.chc.us.ac.com (Stephen M. Shelly):
| ... I need to run
| a command which will return data in column format, I want to extract 
| particular columns into data structures that I can then manipulate.
| I currently have:
| import os
| system("my_command | awk {'print $What $I $ Need'}")
|  i would like to fill a list of objects with this output, but I 
| do not know how......

Awk processes its input in terms of white space separated fields, and
with luck that may work for columns.  Its output puts those fields
back together, though.  If your Python program needs columns, it's
going to do this work itself, so you might as well skip awk and do
the work in python.

    import os   # or posix
    import string
    file = os.popen(my_command, 'r')
    while 1:
        line = file.readline()
        if not line:
            break
        columns = string.split(line)
        # Now you have to sort out which column is which.
    status = file.close()
    if status:
        status = status >> 8
        print my_command, 'exited with abnormal status', status

Incidentally, you must be very careful with the commands you execute
with popen() and system().  They're interpreted by the UNIX shell, and
have the potential to execute any command on the computer, by an
accident of punctuation.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From fw at s.netic.de  Fri Aug 20 17:32:36 1999
From: fw at s.netic.de (Florian Weimer)
Date: 20 Aug 1999 23:32:36 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu> <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <87emgy6suj.fsf@deneb.cygnus.qad.org>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Ah, have you discovered rlcompleter yet? Put this in your ~/.pythonrc:
> 
> import rlcompleter
> readline.parse_and_bind('tab: complete')

And add `del rlcompleter', to remove the symbol from the global namespace
(the hooks will remain active, though).



From indiana at ev1.net  Thu Aug  5 18:48:48 1999
From: indiana at ev1.net (jim moore)
Date: Thu, 5 Aug 1999 17:48:48 -0500
Subject: Stupid Question?
Message-ID: <Bioq3.214$6E5.3247@typ12.nn.bcandid.com>

Is there a bulit in function that will give you N evenly spaced numbers
between X1 and X2.  For example: 10 numbers between 0 and 1 ([0.1, 0.2, 0.3,
...])
It would be similar to MATLAB's linspace.
How about a function similar to range but allowing a floating point step?
Seems like there would be one out there, so I thought I'd check before I
wrote one.
Thanks,
Jim Moore





From p.t at iname.com  Tue Aug  3 05:00:18 1999
From: p.t at iname.com (Peter Torstenson)
Date: Tue, 3 Aug 1999 06:00:18 -0300
Subject: Events with pyKDE 0.8.
Message-ID: <7o6b0i$12ec@enews3.newsguy.com>

I have defined an event as follows.

def mousePressEvent(self,e):
  if e.button()==ButtonState,RightButton:
    self.menu()

Now. this works fine on the mainwidget. But how do I define it to work with
a widget placed on the main widget. Ex, a label.
I have tried to find an explanation in the Qt-manual but with out luck.
There are no examples as far as I can see that illustrates this.

Very greatefull for help

/Peter Torstensson





From dwelton at linuxcare.com  Tue Aug  3 17:41:13 1999
From: dwelton at linuxcare.com (David N. Welton)
Date: 03 Aug 1999 14:41:13 -0700
Subject: Get 'ifconfig' information via Python
Message-ID: <87yafs8ruu.fsf@linuxcare.com>

So, I think I'm getting close...

import fcntl
import IN
import struct

s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)

res = fcntl.ioctl(s, IN.SIOCSIFHWADDR, ??????)

I can't seem to find what goes in ?????.

Is this the right way to be doing things?  I suppose I can always go
pull apart the sources for ifconfig itself, and make a .so that does
what I need, but I get the feeling that I can do what I need from
within Python.

Yeah, I could popen ifconfig, too, but that's ugly.

Downloading python sources now to see what I can see in the fcntl
module...

Thanks,
-- 
David N. Welton, Web Engineer, Linuxcare, Inc.
415.354.4878 x241 tel, 415.701.7457 fax
dwelton at linuxcare.com, http://www.linuxcare.com/
Linuxcare. At the center of Linux.



From bwarsaw at cnri.reston.va.us  Thu Aug 26 17:30:07 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 26 Aug 1999 17:30:07 -0400 (EDT)
Subject: emacs with python mode
References: <37C50DAD.840B67C5@xsoft.co.at>
Message-ID: <14277.45535.576239.46030@anthem.cnri.reston.va.us>

>>>>> "GWG" == Gerhard W Gruber <g.gruber at xsoft.co.at> writes:

    GWG> I'm using Xemacs 21.0 Beta (August 1998) on Win32 and it
    GWG> works quite fine (is there a newer version out there?).

XEmacs 21.1.6 is the latest release I believe; it's what I use on
Solaris, but I haven't upgraded my Windows version in a while, so I'm
not sure if there's an installer for this version.  I'm sure you can
find a 21.1.something lying around, and I'd definitely recommend
upgrading.

You should also be using the latest python-mode.el, version 3.105.  I
don't think the XEmacs packages have been upgraded yet, but you can
download it from http://www.python.org/emacs/python-mode

    GWG> When I'm using Python emacs recognices the mode but it
    GWG> doesn't automatically activate syntax highlighting, even
    GWG> though I activated it in the preferences (same for C/C++ mode
    GWG> as well).

Hmm, maybe you haven't saved your options?  I don't use the menus at
all, but IIRC you want Options->Syntax Highlighting->Automatic and
then save your options.
    
    GWG> Also I was wondering what to do about the tabs.

Personally, I never use tabs anymore, just spaces.  The latest
python-mode tries to be smart about setting the important variables
based on whether the guessed indentation level is equal to a
tab-width.  See the variable py-smart-indentation.

The problem comes, as you've noticed, with files that have an
inconsistent mixture of tabs and spaces.  My recommendation then is to 
set your tab-width to 4, untabify the buffer (C-x h M-x untabify RET), 
then save and revisit the buffer (so tab-width goes back to 8 and
py-smart-indentation can do it's thing).
    
Hope that helps,
-Barry



From sshelly at pcc.net  Sat Aug 28 08:22:51 1999
From: sshelly at pcc.net (Stephen)
Date: Sat, 28 Aug 1999 07:22:51 -0500
Subject: creating a list of class instances from file / stdin (AGAIN)
Message-ID: <37C7D49B.CA7DEAE1@pcc.net>

First: thanks to Thomas Wouters and Donn Cave for the assist

I currently have
mb_program = os.popen("my_command | awk {'print $1\" \" $8\" \"$7\"
0\"'}", "r")
for line in mb_program.readlines():
        mb_list.append(string.split(line))

I have a class:
class MyClass:
    def __init__(self,field_one,field_two,field_three)
        self.field_one = field_one blah, blah

  What is the syntax to do:
for line in mb_list:
    create an instance of this class and append it to a new list????
Can I combine the mb_list.append statement above to create instances of
the class while reading the input?
The poroblem I am trying to solve essentially involes a table with three
fields, the first two of which can be duplicates.
Like I said in previous emails, I am pretty new to python, but the news
group is really much more helpful to us newbies than the other languages
I have tried..

PEACE




From spamfranke at bigfoot.de  Wed Aug 18 13:24:23 1999
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Wed, 18 Aug 1999 17:24:23 GMT
Subject: Why no "def A.f (...)"?
Message-ID: <37bae4c3.98260320@news.omnilink.de>

Are there sensible reasons (performance, anomalies..) why 
function definition is restricted to creating new identifiers 
in the current namespace?

An extended syntax 

  def A.B.f (self, x):
    <body>

with semantics like

  setattr (A.B, "f", lambda self, x: <body>)

would be the most straight forward solution for my current problem,
creating classes at runtime. -- OK, of course there are *lots* of 
other ways to do this, but it would also make the language a 
little bit more homogenous IMHO.

Does anyone else see a use for this?

Stefan




From jkmuir at uniserve.com  Fri Aug 27 20:43:04 1999
From: jkmuir at uniserve.com (Justin Muir)
Date: Fri, 27 Aug 1999 19:43:04 -0500
Subject: A scrollable Frame or Canvas widget
Message-ID: <37C73097.8165F42C@uniserve.com>

Hello!  I've  been experiencing total embuggerance in attempting to
make a scrollable Frame or Canvas widget.

Essentially I'm trying to place a schwack of Labels into a window that
can be scrolled.   I can get the scrollbar widget to show up at the
bottom of the screen but the arrows are just back to back with no
scrollbar and no functionality!

Thus far, without success, I've tried a Frame and a Canvas with the
scrollbar.   I instantiate all the labels in a grid layout and then add
the Scrollbar.   All good as it shows up... but now, how do I make it
function?

This is the general idea of what I'm attempting:

class windowWithABunchOfLabels:
            def __init__(self, master):
                    self.c = Canvas(master)
                    self.c.config(scrollreigon=self.c.bbox(ALL))
                    self.c.pack()

                    for x in range(30):
                        for y in range(20):
                            Label(c, text='some stuff',
relief=RIDGE).grid(row=y, col=x)

                    self.horizontalScrlBar = Scrollbar(c,
orient=HORIZONTAL, command=?????)
                    self.horizontalScrlBar.grid()


I've pored over the Tkinter docs and cannot come to any reasonable
solution.
..........I'm kinda thick so be patient :-)


Could anyone out there also clear up the mechanics of the different
callbacks/commands invoked by the arrows and the scrollbar?  I read
Frederick Lundh's Tkinter guide and it still wasn't too clear.


Thanks!

Justin.




From p.t at iname.com  Sun Aug  1 06:57:44 1999
From: p.t at iname.com (Peter Torstenson)
Date: Sun, 1 Aug 1999 07:57:44 -0300
Subject: How to find out the pid of a running program?
References: <7nucu2$11ko@enews1.newsguy.com> <37A31E4B.2CD53D40@ibm.net>
Message-ID: <7o1989$pq7@enews2.newsguy.com>

> Peter Torstenson wrote:
> >
> > I'd like tknow if there is a way of getting the pid of a program
running,
> > using python commands?
>
> Use the getpid() function from the posix module which is imported
> automatically when you import the os module.  Don't import the posix
> module directly.  Do it via the os module since this will afford you
> code better portablility.  Refer to the Python Library Reference for
> more info.

I think I didn't express myself clearly!
I want to find out the pid of another program than the one I give
the command from!
Ex. the pid of the running pppd.
The getpid() function returns the pid of the python program it self as far
as I understand.







From gherman at my-deja.com  Thu Aug 12 08:23:51 1999
From: gherman at my-deja.com (Dinu C. Gherman)
Date: Thu, 12 Aug 1999 12:23:51 GMT
Subject: Zope for python.org?
Message-ID: <7ouecd$9m$1@nnrp1.deja.com>

Just a few quick questions...

Would it be desirable to turn python.org into a Python "portal
site" using the increasingly popular Zope product?

Would that show a strong "cross-commitment" between Python and
Zope and is it desired to show that?

Would that allow for a multi-language, multi-editor, better
integrated, easier to manage "grand" Python site, preventing
"spin-offs" like non-English copies?

Would that be a big problem for mirroring sites?

Dinu

................................................................
"An average of more than 15 % of adults in 12 industrialized
countries are functionally illiterate; in Ireland, the United
Kingdom and the United States, the rates are over 20 %."

  (The State of the World's Children 1999,
   UNICEF, http://www.unicef.org/sowc99)




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From chad at vision.arc.nasa.gov  Mon Aug 16 19:44:37 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Mon, 16 Aug 1999 23:44:37 +0000
Subject: C's isprint() concept?
References: <000201bee78f$6bbf6ec0$f22d2399@tim> <7p9d0v$m9c@dfw-ixnews11.ix.netcom.com>
Message-ID: <37B8A265.8A9EAC69@vision.arc.nasa.gov>

Aahz Maruch wrote:

> Ah.  Interesting.  I'm using re.sub because we're still on 1.5.1, but
> we're supposed to move to 1.5.2 Real Soon Now.

Just out of curiosity, what is the biggest source of the delay from
switching to 1.5.2?  Is it a matter of "if it ain't broke...", or do the
changes warrant enough work (rewriting code, updatin libraries,
etc.) that it is a big job?  I'm writing apps, and currently rely on
some 1.5.2 behaviors, so I'm curious as to how difficult it may be
for customers to upgrade if need be.

Chad Netzer
chad at vision.arc.nasa.gov





From donn at u.washington.edu  Thu Aug 19 14:29:40 1999
From: donn at u.washington.edu (Donn Cave)
Date: 19 Aug 1999 18:29:40 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7ro9ns.h6v.jblaine@shell2.shore.net> <37bc2ee5@cs.colorado.edu> <slrn7rog4u.hri.jblaine@shell2.shore.net>
Message-ID: <7phiek$sn4$1@nntp6.u.washington.edu>

Quoth jblaine at shell2.shore.net (Jeff Blaine),
...
| I know what version of Python I am running, and I know where the docs
| for that version are.  Using Python 1.5.x and expecting a book which
| is several years old (and based on 1.4) to be your source of
| day-to-day reference seems pretty silly if you ask me.

Worth pointing out it wouldn't be a showstopper, though.  There would
be some rough spots (changes in exception handling come to mind), but
overall I think the PostScript (library, reference, tutorial) doc from
1.1 would be pretty much on target (happens to be the first version I
remember.  This core stability is itself a virtue of Python, but I wonder
if it has inhibited the development of the documentation.)  Now if you
wanted to produce a comprehensive article on the features missing from
Python, you certainly would be missing a bet not to refer to current
documentation, but that's an unusual objective.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From bennettt at am.appstate.edu  Thu Aug 12 19:48:59 1999
From: bennettt at am.appstate.edu (TMGB)
Date: Thu, 12 Aug 1999 19:48:59 -0400
Subject: internet programming with python
References: <37AF678E.CC8E165B@netscape.com> <37B1AB33.1C81BDEA@am.appstate.edu> <37B20DA3.3E3ABD98@malcolmson.com>
Message-ID: <37B35D6A.E148A60E@am.appstate.edu>

Andrew Malcolmson wrote:
> 
> I also have it "Internet Programming with Python" and found it very good, fun
> to read, and probably a better book for a Python beginner than "Learning
> Python".  I hope it gets updated.
> 

> However, are you saying that there is something in it about using Python with
> Netscape Enterprise Server?  I didn't see anything there.

Check "Chapter Twelve: Embedding Python.." pages 425-443 is almost
completely about embedding Python with the Netscape Enterprise server
through Netscape's NSAPI.

Thomas


> 
> TMGB wrote:
> 
> > I just picked it up at Barnes and Noble where it retailed for $34.95 US
> > and is discounted to $17.48 US.  I feel sure that you can probably order
> > it from http://www.barnes&noble.com (not sure about URL).  The ISBN is
> > 1-55851-484-8 published by M&T Books a division of MIS:Press,
> > INC(http://www.mandt.com).  I'm not sure but I think some references are
> > to libraries that have now been updated or replaced but in general it
> > has been a very helpful introduction.  Especially interesting (to me),
> > if you have Netscape Enterprise Server, is embedding Python into
> > Netscape Server NSAPI.
> >
> > Thomas
> >
> > Vijay Mallela wrote:
> > >
> > > Hi,
> > > I am looking for the book "Internet programming with python". I need it
> > > at the earliest..
> > > I checked out amazon and other stores and they dont have it.. does any
> > > one know where I can get the book
> > > Thanks
> > > vijay
> >
> > --
> > ----------------------------------------------------------------------
> > Thomas McMillan Grant Bennett           Appalachian State University
> > Computer Consultant II                  University Library
> > bennettt at am.appstate.edu
> > http://www.library.appstate.edu/admin/
> > Voice:  828 262 6587                    FAX:    828 262 3001
> >
> > Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
> > operating system that was originally coded for a 4-bit microprocessor.
> >  - Chris Dunphy     Boot Magazine

-- 
----------------------------------------------------------------------
Thomas McMillan Grant Bennett           Appalachian State University
Computer Consultant II                  University Library
bennettt at am.appstate.edu               
http://www.library.appstate.edu/admin/
Voice:  828 262 6587			FAX:    828 262 3001

Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
operating system that was originally coded for a 4-bit microprocessor.
 - Chris Dunphy     Boot Magazine



From fredrik at pythonware.com  Tue Aug 17 05:18:46 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 17 Aug 1999 09:18:46 GMT
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org> <14265.5096.537800.149227@buffalo.fnal.gov> <37B91E9A.2451936C@w3.org>
Message-ID: <001301bee891$82fccc60$f29b12c2@secret.pythonware.com>

(posted and mailed, as usual.  or to be precise, mailed and mailed ;-)

Dan Connolly <connolly at w3.org> wrote:
> >  > I hope the os module will be deprecated in future releases.
> > 
> > I don't think there is much chance of this happening, it it too widely
> > used.
> 
> How about attempting to change that trend? i.e. how about
> in stead of documenting os.getpgrp() with an availability
> caveat, the os module documentation only lists the stuff
> that's really expected to be portable, and has a note
> at the bottom ala:
> 
> the os module may contain other implementation-defined
> objects; for example, on posix platforms, the contents
> of the posix module are shadowed in the os module.
> [... list the other cases ...]

sounds reasonable.

(in other words, it's mostly a documentation
problem, rather than a design problem...)

(Fred?  Guido?)

</F>




From Sven.Drescher at dlr.de  Mon Aug 16 06:09:51 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Mon, 16 Aug 1999 12:09:51 +0200
Subject: an other thread to embed Tk???
Message-ID: <7p8n7n$nsi@news.go.dlr.de>

Hi!

I want to embed Python to C++, because I need a GUI to interact with my
C-program. But so easy like in my book "Programming Python" it isn't.

I load the Tkinter-module, but I have the following error message:
"Fatal Python error: PyThreadState_Get: no current thread"

I used VC++6.0EE to compile and link it.

I think I must create a new thread for my Python-GUI-code, but I don't know
the way I have to do it.

Thanks for hints and sites, where I can get information!

Sven






From jstevens at bamboo.verinet.com  Fri Aug 13 23:18:19 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: Sat, 14 Aug 1999 03:18:19 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com>

On 14 Aug 1999 02:39:10 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>On 13 Aug 1999 20:05:12 -0700,
>	John W. Stevens <jstevens at basho.fc.hp.com> wrote:
><snip>
>>
>>Thanks for the two missing magic characters (@$, and =>).  I always
>>have to look them up.
>
>Learn some perl then. @ is pretty fundamental. $ is pretty fundamental.
>What happens when you tell a scalr to be an array is pretty fundamental.

Oh, I *KNOW* the semantic.  I just can't remember the syntax.

As for calling the conversion of a scalar to an array "pretty
fundamental". . . can you say "Extreme Exageration"?

Tom tells me that Perl doesn't exclude non-computer scientists,
then you guys start throwing around references to references. . .

:-)

>Only if you don't know perl.

No, it is confusing to the average programmer.  If it isn't
confusing, to you, then you've spent a *LOT* of time learning
and using Perl.

>$dict is a scalar. %dict is a hash. In that
>example a scalar was used. 

Yes. . . is it a hash, or a scalar?  If it is a scalar, why
is it called dict?  If it is a hash, then why is it prefixed
by $?  If this is a reference instead of a scalar, then why
doesn't it have it's own special prefix character.  ;->

>At least in the discussions I have with people about python I don't
>complain about python syntax I don't understand I just ask what it means.

I am not complaining about syntax I don't understand.  I am complaining
about syntax that is unecesarily difficult to learn, remember, and use.

>Instead of calling it confusing people who actually want to learn something
>ask what it means, and why it is so. You obviously don't want to
>learn but only to criticise. That's fine but it would be better to just
>post to the python group...

What, you believe that you are a mind reader?

You are wrong.  I know what the syntax means.  I criticize because
it is indeed confusing, and difficult to remember.

>You can write Perl programs that resemble sed, or awk, or C, or Lisp, or
>Python. This is Officially Okay in Perl culture.

You cannot write Perl that resemble Python.  You are required to 
use curly braces as block delimiters.

John S.



From tim_one at email.msn.com  Sat Aug 21 21:25:25 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 21 Aug 1999 21:25:25 -0400
Subject: More random python observations from a perl programmer
In-Reply-To: <37bec222@cs.colorado.edu>
Message-ID: <000101beec3d$36204fc0$cb2d2399@tim>

[posted & mailed]

[Tom Christiansen]
>:>    You can't use "in" on dicts.  Instead, you must use

[Neil Schemenauer]
>: In the Python philosophy, every expression should have an obvious
>: meaning.  "in" in this case does not.  Perl has a different
>: philosophy here.

[Tom Christiansen]
> To me it would be clear that it's a wasteful linear search of the
> values, just as in a list it's a wasteful linear search of the
> elements.

Python doesn't mind if you want to do something (theoretically) inefficient,
and e.g.

    x = [3, 5, 11]
    if 5 in x:
        print "yes"

does print yes (darned fast, too <wink>).  So the ambiguity of "thing in
dict" doesn't resolve itself on this basis.  Many people also view sequences
as mapping an index set to the sequence elements, and under that view the
current "thing in sequence" isn't asking whether thing is in the index set;
so "for consistency", then, "thing in dict" would have to refer to
membership in dict.values().  Finally, since you can already ask about key
membership directly via dict.has_key(thing), it would be un-Pythonic to
introduce a second spelling ("there's only one (most obvious) way to do it"
is a Python goal).

[on what's a local]
> I understand.  But it's well, surprising that it's all context-
> dependent.  Not fun to look for.

But there's nothing to look *for* <wink>.  That Python does not have
Scheme-like lexical closures, or dynamic scoping either, means that each
unqualifed name is either purely local or global; there are no other
possibilities to worry about, and globals used on the LHS of an assignment
have to be explicitly declared "global" *in* every function they're so used.
So global-vs-local is a purely local static property of a function's source
text, and no context outside of a function's body need be looked at.  It's a
non-issue in practice.

IMO Python's rules wouldn't work nearly as well in the presence of nested
lexical scoping, so I've got mixed feelings about introducing lexical
closures.  Most closures are "tiny & simple" enough that the
default-argument trick is semantically adequate albeit syntactically clumsy.

> ... but I've never understood why
>
>     a += b
>
> isn't immediately obvious as
>
>     a = a + b

That would be OK by me, but it would preclude some very natural uses for the
shorthand syntax.  Mentioned before e.g. that the NumPy people would love
for

    a += b

to update matrix a in-place ("a" and "b" may each consume many dozens of
megabytes, and in-place modification can buy order-of-magnitude runtime
efficiencies when mucking with huge objects).

Guido isn't opposed to += & friends, they simply hasn't been implemented
yet.

>: As I mentioned earlier, everything is a reference.  You just have
>: to know what is mutable and what is not.  Lists, obviously from
>: your example, are.

> Alas, I'm pretty sure that this isn't how people think unless
> they're trained to do so.
>
> There's a reason that in Perl most people still use
>
>     @a = (1,2,3);
>
> not
>
>     $a = [1,2,3];
>
> It's because @a is a first-class array, and $a merely a reference
> thereto.

The majority of Perl dabblers (as opposed to serious Perl programmers, but I
don't know many of those) I know avoid the latter form becase Perl's flavor
of references baffles them.

I don't think it's the "referenceness" per se, but the spelling that hangs
them up -- after the above, I'm not sure you can still sympathize with
someone asking "now what the heck is the difference between '$$a' and
'@$a'?!".  I agree it's perfectly logical, but *I* still hear it a lot <0.5
wink>.

The other thing I see a lot is

    @a = [5, 6, 7];

which is a very tempting mistake for people coming from other languages
(like, say, Python).  This doesn't create the array they expected, but a
1-element array containing a reference to the array they were expecting, and
e.g. now you have to get at the 2nd element via ${$a[0]}[1].  Last time I
tried that, "perl -w" let it slip by silently, too.  As you've noted
eloquently before, array-vs-scalar context is a "deadly sin" in Perl, and
I'll add that the introduction of references only made that aspect of the
language even touchier (now things that "look like" arrays can actually be
scalar).

> Copying things not producing a copy is well, not what non-serious
> programmers immediately think of.

Patience, here.  People will repeat this until it sinks in <wink>:
assignment never copies in Python.  It just binds a name to an object.  This
is something many (by no means all) people do struggle with at first, but I
think it's because it's *simpler* than what they're expecting, and it takes
some time before they stop trying to apply the overly convoluted mental
models they bring with them from other languages.  Persist, and the light
will eventually dawn!

> Why can python copy sequences of chars (1-byte strings) trivially:
>
>     s = "this stuff"
>     t = s
>
> But as soon as you have a sequence of, oh, integers for example, you
> can't use the same set-up.  Very odd.

Nope, that example and

    s = [1, 2, 3]
    t = s

are exactly the same.  In the first case, s and t end up naming the same
string ("s is t" returns true after that snippet); in the second case, s and
t end up naming the same array; and after

    s = anything_whatsoever
    t = s

s and t are both names for whatever anything_whatsoever computed.
Absolutely no exceptions.  The semantics of argument-passing work exactly
this way too: the formal argument names are simply bound to the actual
arguments passed by the caller, as if by a sequence of assignment
statements.  Ditto for
    from module import thing
In that case the local name "thing" is attached to the object module.thing.
Ditto for
    import module
There the local name "module" is attached to the module object.

Nothing is copied, no how, no way, never ever.  If you *want* a copy (of an
list, string, or anything else), you have to explictly make one.

Now the interesting question <wink> is why you *think* the string in your
example got copied!  Your mental model is conjuring up an inconsistency
where none actually exists in the language, so "the answer" to that one lies
not in Python but in what you're bringing to it.

Another interesting question is whether by-value or by-reference semantics
are more useful more often.  OO languages almost always pick the "by
reference" answer, and for good reasons both theoretical and practical.
Although sometimes it would be a lot more convenient if e.g. the "Tom
object" could get copied and go to 3 conferences simultaneously <wink>.

>:> ...
>:> GOTCHA: (medium)
>:>    All ranges are up to but *not including* that point.

The intended mental model is that indices point *between* elements, with 0
pointing "to the left" of the leftmost element, and -1 pointing to the left
of the rightmost element.  Let that sink in, and then everything else is
obvious.

A pair of endpoints could have 4 interpretations, depending on whether each
endpoint is inclusive or exclusive; that's why viewing slice notation as
naming the endpoints is a bad idea.  There are no surprises when viewing it
as intended.

Perl worms around a similar problem in "substr" and "splice" by using
offset+length coordinates, although a negative length (in at least "substr";
"split" too?) is special-cased to mean what it means in Python instead.
Unless you think of

    substr($s, 3, -1);

as meaning "all the characters starting at index 3 up to but not including
the character at position -1" <wink>.

Python's slice notation is Major Goodness, Tom!  It's also Major Goodness
that the same notation works on arrays, strings, tuples and user-defined
sequences.

>: This is explained quite well in the tutorial.  It makes a lot of
>: sense once you get used to it.

> As before, "once you get used to it" is the key point here.

Python isn't particularly concerned about being obvious at first glance
(*no* computer language could be deluded enough to claim that!); it's more
concerned with notations that, once mastered, are very hard to forget or
misuse by mistake.  Don't think "obvious at first", think "obvious forever
after".  And I agree the singleton tuple notation is a wart.

>: Again with the references. :)  Don't think references.

> You can't avoid it.

It *is* a lot like giving up a thirty-year-old alcohol addiction <wink>.  I
think "objects" in Python (Java too, for that matter), almost never
"references".  Objects have names, and that's about it; thinking about
pointers doesn't do much good for you in Python unless you're mucking with
its implementation.

> ...
> Huh?  How would I use a dictionary (don't you get tired of such a long
> and non-intuitive word?) to do this:
>
>     @a[3,7,4] = ("red", "blue", "green")
> or
>     @a[3,7,4] = @a[2,9,10];

Python doesn't have scatter/gather subscripting (note that most of the
arguments you're having about this here are due to "slice" meaning different
things in Python and Perl).  You can do the above in one line by using "map"
and operator.setitem, but a sane person would simply write a loop today, or,
if they did this a lot, capture that loop in a generalized function.  Python
may have gather/scatter subscripting in the future (the NumPy people are
keen for it; but iirc they're the only ones pushing it).

...
>: del is a statement.  It has to be so it can remove variable
>: bindings.

> I *really* don't undertsand this. Why does Python have so many
> "statements" compared with C or Perl?

To the small extent that it does (or don't you count "for" as distinct from
"foreach", or "while" from "until", or "if" from "unless"?), it's because
Python's lineage traces more directly to the stmt-oriented Algol than to the
expression-oriented C.

Why do Perl & Python both have so many statements compared to, say, Icon or
Lisp?  Why do they both have so few statements compared to, say, Fortran or
COBOL?  Is there a *real* reason Perl has the precise set of statements it
has, other than that it suited Larry Wall's taste at the time?

Python's "del" needs to be a statement because, e.g., as a function call

    del(a[3:7])

couldn't work, and having used the slice notation consistently everywhere
else, backing off to e.g. splice(a, 3, 4) wouldn't fit with the rest of the
language.

Python's "print" doesn't *need* to be a statement, but is one simply because
it's a common operation and Python requires "()" in function calls -- that
is, it's a statement purely for convenience.

Etc.

> ... I can ... without needing to sacrifice an entire *statement*
> in the grammar for this!

Are statements a scarce resource we need to preserve for future generations?
Python's parser is generated from a formal grammar specification; the latter
is in a file that's 90 lines long, a third of which is blank lines and
comments, and where precedence levels and associativity are fully spelled
out by the grammar structure (not by yaccish shorthands).  IOW, Python's
grammar is exceptionally small for a procedural language as-is (contrast
perly.y).

BTW, when I was writing Python's Emacs mode, Bill Mann worked down the hall
plugging away at Perl mode -- I hadn't before realized man's capacity for
suffering <wink>.

> ...
> There must be something important that I'm not understanding
> here, and I'd like to.

Expressions-vs-statements is a design rift that began no later than Fortran
vs LISP.  Neither Python nor Perl is at either extreme of that spectrum;
both are firmly in the squishy middle.  And only Lutherans suspect there's
something important differentiating them from Episcopalians <wink>.

once-weekly-vs-four-times-a-month-ly y'rs  - tim





From evan at tokenexchange.com  Wed Aug 25 11:34:12 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Wed, 25 Aug 1999 10:34:12 -0500
Subject: xanadu and python
References: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com> <14276.2083.275770.273576@dolphin.mojam.com> <slrn7s82hf.g2c.kaz@ashi.FootPrints.net>
Message-ID: <52Uw3.40$2y4.43051@news-dal.corridex.com>

You should be able to reach http://udanax.xanadu.com if you're interested.
That 500 Server Error certainly is odd, though.

Kaz Kylheku <kaz at ashi.FootPrints.net> wrote in message
news:slrn7s82hf.g2c.kaz at ashi.FootPrints.net...
> On Wed, 25 Aug 1999 10:15:00 -0500 (CDT), Skip Montanaro <skip at mojam.com>
wrote:
> >Well, I get:
> >
> >    500 Server Error
> >
> >    The hard transfer limit for this user has been reached
> >
> >Not sure what that has to do with either Python or Xanadu...
>
> Nothing. It has to do with the Slashdot effect. The story about Xanadu
> being open sourced hit Slashdot last night.





From paul at prescod.net  Sat Aug 28 06:39:15 1999
From: paul at prescod.net (Paul Prescod)
Date: Sat, 28 Aug 1999 10:39:15 GMT
Subject: Python Type System: An idea for unification
References: <37c72e93.1411973891@news.triode.net.au>
Message-ID: <37C7BC53.3D1D8B8C@prescod.net>

John (Max) Skaller wrote:
> 
=> One of the things that is often requested for python is that
> builtin types be class instances. Viper does _not_ do this:
> AFAICS it is theoretically impossible (and  less useful than
> the technique described below).

Could you please describe why this is theoretically impossible? Thanks,

 Paul Prescod



From rjroy at takingcontrol.com  Wed Aug 25 09:21:16 1999
From: rjroy at takingcontrol.com (Robert Roy)
Date: Wed, 25 Aug 1999 13:21:16 GMT
Subject: xanadu and python
References: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com>
Message-ID: <37c3eca5.54328515@news1.on.sympatico.ca>

On Wed, 25 Aug 1999 08:03:19 GMT, "Fredrik Lundh"
<fredrik at pythonware.com> wrote:

>I doubt anyone have missed that Ted Nelson's Xanadu,
>the original hypertext system, has finally been released
>as open source:
>
>    http://www.udanax.com/
>
>what Python has to do with it?  well, go to the site and
>find out for yourself ;-)
>
></F>
>
oops spoke too soon
the url did not work for me so I went to xanadu.com which referred me
to the mirror. udanax.com isn't fully propagated yet.





From bwarsaw at cnri.reston.va.us  Thu Aug 19 14:45:36 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 14:45:36 -0400 (EDT)
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
	<19990819174800.D5227@xs4all.nl>
	<37bc313b@cs.colorado.edu>
Message-ID: <14268.20688.686323.933935@anthem.cnri.reston.va.us>

>>>>> "TC" == Tom Christiansen <tchrist at mox.perl.com> writes:

    TC> No, I'm sorry, the expressivity of "break LABEL" and "continue
    TC> LABEL" are much clearer than the cicumlocutions necessary for
    TC> exceptions.

Plus, they'd probably be faster.  This is a good point: being able to
label breaks and continues a la Java would be useful.  I sometimes
miss them in Python but the workarounds are easy enough.
    
-Barry



From pmagwene at uchicago.edu  Tue Aug  3 16:36:07 1999
From: pmagwene at uchicago.edu (Paul M.)
Date: Tue, 3 Aug 1999 20:36:07 GMT
Subject: Dislin is a nice plotting package
References: <Pine.LNX.4.10.9908021807550.12462-100000@us2.mayo.edu>
Message-ID: <FFwov4.CEA@midway.uchicago.edu>

I must concur with Travis' assessment.  I've been using Dislin with
Python for the last 6 months or so, and have been very pleased with
the quality of plots I've been able to produce (I used many DISLIN
generated plots in my dissertation).

I written a very rudimentary object-oriented wrapper around the basic
DISLIN API.  I wanted to clean it up before releasing it (along with a
package I've been doing some multivariate statistics with) but perhaps
others are interested as well.

Is this of general interest?  Has somebody else already written
something similar?  I'm in the middle of moving cross-country right
now, but in a few weeks I should be able to throw my code and some
basic documentation onto a web page.


--Paul Magwene
pmagwene at midway.uchicago.edu



Travis Oliphant <Oliphant.Travis at mayo.edu> wrote in message
news:Pine.LNX.4.10.9908021807550.12462-100000 at us2.mayo.edu...
> A users experience with plotting under Python:
>
> I know plotting has been discussed on this list before.  Many people
> indicate that the lack of quality plotting is preventing them from
using
> Numerical Python as a complete replacement for IDL or MATLAB.
>
> I have been using gist for the past several months and found it to
be
> excellent for investigative plotting (except I miss axes on surface
> plots).  The plot quality and flexibility with gist is excellent (it
is
> particularly easy to use for me with the matlab-like plotting
commands I
> implemented)
>
> However, recently I needed to produce plots with legends on them so
I
> turned to DISLIN.  I had seen DISLIN before and even had it
installed but
> it looked pretty complicated to use so I hadn't tried it since I
wasn't
> desperate for good plots.  Well, I've now tried it and I'm
impressed.
> It is a COMPLETE plotting system.  After a bit of reading to
understand
> the plotting paradigm (it is a very powerful but non-obvious system)
I can
> now make any plot I need with the system, and configure it to my
hearts
> content.
>
> There are even quick plotting functions defined so that I can make
> surface plots (with axes) and color plots with an automatic color
bar,
> with a single plot command.  I can even use dislin and gist together
which
> works very well.
>
> An object-oriented wrapper around the quick-plots so that all the
> properties could be set easily would make dislin quite phenomenal
for use
> with Python.
>
> I don't mean this post to discourage work on the fine plotting
efforts
> underway with PIDDLE, GRAPHITE, and Snow.  These packages will only
> improve the plotting choices under Python.  I just wanted to express
my
> pleasure at using DISLIN under Python and encourage others who may
be
> frustrated about their ability to produce quality output now or who
> may be holding back from adopting Python as their interactive
> analysis environment to check out DISLIN and see for yourself if it
will
> serve you.
>
> -Travis
>
> P.S. The dislin homepage is
>
> http://www.linmpi.mpg.de/dislin
>
>
>




From gmcm at hypernet.com  Fri Aug 27 22:33:16 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 27 Aug 1999 21:33:16 -0500
Subject: Why do we call python a scripting language?
In-Reply-To: <37c726c4.2148316@news.netset.com>
Message-ID: <1276318766-38718063@hypernet.com>

Someone wrote:

> But what I find most interesting is that I havent seen much evidence
> that any one uses it.  Is any one using freeze or some derivative or
> work alike?  Most things I find are written for users of python, and
> are just distributed as source.

What you see here is Pythonistas talking to Pythonistas; you know, my 
algorithm is more macho than your algorithm. Lots of people use 
freeze & friends to distribute stuff, but they're distributing apps, 
not Python.

- Gordon



From fdrake at acm.org  Thu Aug 19 11:37:45 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 19 Aug 1999 15:37:45 GMT
Subject: whrandom documentation (was Re: Unsorting(randomizing) a sequence)
In-Reply-To: <14268.6727.737269.35392@anthem.cnri.reston.va.us>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<14265.47748.241386.291144@buffalo.fnal.gov>
	<wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
	<m3672dwojr.fsf@pcrm.win.tue.nl>
	<14266.49377.51221.305982@dolphin.mojam.com>
	<14266.59802.56315.731454@bitdiddle.cnri.reston.va.us>
	<14268.6727.737269.35392@anthem.cnri.reston.va.us>
Message-ID: <14268.9417.310224.324899@weyr.cnri.reston.va.us>

Barry A. Warsaw writes:
 > In that case, shouldn't the documentation for the whrandom module
 > recommend this usage?

  Probably.  It looks like the documentation for random and whrandom
contain annoying interdependencies (random should not point to
whrandom to describe its interface; whrandom should claim that it
supports that part of the random interface).


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From greg.ewing at compaq.com  Mon Aug 23 20:14:14 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 24 Aug 1999 12:14:14 +1200
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> 	<m3vham44g5.fsf@atrus.jesus.cam.ac.uk> 	<14257.54137.72005.705719@anthem.cnri.reston.va.us> <37b6f1c2.347828751@news.triode.net.au>  <37B82032.E7A11398@starmedia.net> <slrn7rpbk8.7pl.neelk@brick.cswv.com>
Message-ID: <37C1E3D6.CD835CD1@compaq.com>

Andrew Csillag <andrew at starmedia.net> wrote:
>
>Also, you may want to look at how the SmallEiffel compiler does it's
>type inferencing...  
>(and obviously the souce code is available to give you an implementation
>>to look at -- written in Eiffel).

Are you sure about the source code? Last time I looked
at SmallEiffel, they didn't give you the Eiffel source,
only the generated C code.

Has this changed? If so, it's very interesting.

Greg



From bennettt at am.appstate.edu  Wed Aug 11 12:56:19 1999
From: bennettt at am.appstate.edu (TMGB)
Date: Wed, 11 Aug 1999 12:56:19 -0400
Subject: internet programming with python
References: <37AF678E.CC8E165B@netscape.com>
Message-ID: <37B1AB33.1C81BDEA@am.appstate.edu>

I just picked it up at Barnes and Noble where it retailed for $34.95 US
and is discounted to $17.48 US.  I feel sure that you can probably order
it from http://www.barnes&noble.com (not sure about URL).  The ISBN is
1-55851-484-8 published by M&T Books a division of MIS:Press,
INC(http://www.mandt.com).  I'm not sure but I think some references are
to libraries that have now been updated or replaced but in general it
has been a very helpful introduction.  Especially interesting (to me),
if you have Netscape Enterprise Server, is embedding Python into
Netscape Server NSAPI.

Thomas

Vijay Mallela wrote:
> 
> Hi,
> I am looking for the book "Internet programming with python". I need it
> at the earliest..
> I checked out amazon and other stores and they dont have it.. does any
> one know where I can get the book
> Thanks
> vijay

-- 
----------------------------------------------------------------------
Thomas McMillan Grant Bennett           Appalachian State University
Computer Consultant II                  University Library
bennettt at am.appstate.edu               
http://www.library.appstate.edu/admin/
Voice:  828 262 6587			FAX:    828 262 3001

Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
operating system that was originally coded for a 4-bit microprocessor.
 - Chris Dunphy     Boot Magazine



From gmcm at hypernet.com  Tue Aug 31 11:51:04 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Tue, 31 Aug 1999 10:51:04 -0500
Subject: Python speed and `pcre'
In-Reply-To: <oqzoz89hrb.fsf@titan.progiciels-bpi.ca>
Message-ID: <1276011696-57188264@hypernet.com>

Fran?ois Pinard writes:

> After having translated some code (not big, but not small) from Perl
> to Python, I discover it runs ten times slower.  I did not learn to
> profile yet, and I am not sure it would help, after having read that
> profiling gives call counts, but no elapsed time.  Is that right?

No.
 
> My intuition tells me that the amount of regular expression matching
> might provide an explanation.  Here is my set of _hypotheses_ (I'm
> not sure):
> 
> * Perl keeps compiled all /REGEXP/ not using string interpolation,
> * Python cache for compiled REGEXP is less virtuous than I imagined.

Just the last one used, I think.

[wants to cache the re.compiled regex, but keep the regex string near 
its use]

Here's one way of caching (note that I don't do anything like this - 
I just put all my patterns at the top and compile them right away. 
The intent of a regex is often better communicated through giving it 
a meaningful name, rather than staring at the regex itself):
---------------------------------------
import re

ccache = {}
def ccompile(regex):
  if not ccache.has_key(regex):
    ccache[regex] = re.compile(regex)
  return ccache[regex]

def cmatch(regex, s):
  return ccompile(regex).match(s)

def csearch(regex, s):
  return ccompile(regex).search(s)

s = "This little piggie had Spam, This little piggie had None"

mo = csearch('piggie', s)
print mo.group(), 'at', mo.start()
mo = csearch('piggie', s[mo.end():])
print mo.group(), 'at', mo.start()
--------------------------------------------
Now had I been using explicitly compiled regexes, I would not have 
had to slice s, since the compiled regex object lets you specify 
start and end (without a copy being made). Yes, I could add that to 
this code, but why bother?

- Gordon



From tdixon.no at spam.fwi.com  Tue Aug  3 11:48:40 1999
From: tdixon.no at spam.fwi.com (Timothy Dixon)
Date: Tue, 03 Aug 1999 15:48:40 GMT
Subject: Python in web page
Message-ID: <37a70f28.96313821@midpoint>

Is it possible to host a Python application in a web page?  TCL has a
plug-in for IE and Netscape; is there something similar for Python?




From nathan at islanddata.com  Tue Aug 24 19:31:34 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Tue, 24 Aug 1999 16:31:34 -0700 (PDT)
Subject: MySQLdb python module dumps core!
Message-ID: <XFMail.990824163134.nathan@islanddata.com>

I've run the MySQLdb module successfully on two different Suns running
Solaris 2.6.  Just put it on a new machine running 2.7 and it dumps core
on errors!  On the working host:

  File "/usr/local/lib/python1.5/site-packages/MySQLdb.py", line 245, in
__init__
    self.db = apply(connect, (), kwargs)
_mysql.OperationalError: (1045, "Access denied for user:
'erlogs at localhost' (Using password: YES)")

Which is what *should* happen.  If I log in correctly, all works fine, but
I really need to be able to catch these exceptions rather than dumping
core with a bus error :)

Python 1.52 and mysql-3.22.25-sun-solaris2.6-sparc binary install.  Any
ideas, please share!  Thanks.


WARNING: message cross-posted to two mailing lists.  Reply with care.



----------------------------------
Nathan Clegg
 nathan at islanddata.com





From bryn at mds01.itc.com.au  Fri Aug  6 01:23:44 1999
From: bryn at mds01.itc.com.au (B Kingsford)
Date: Fri, 6 Aug 1999 15:23:44 +1000
Subject: Python, Embedding Q's
Message-ID: <7odrn8$qf2$1@wyrm.its.uow.edu.au>

hi all

im embedding Python in multi-threaded server, and just had a
couple of q's regarding compiling files, cache, etc. if any help
or thoughts could be offered, please do.

I wish to use Python's built-in threading of the interpreter,
aswell as Python's memory management [for portability], will
this have any major, adverse performance effects?
Does Python handle which thread does what?
would it be better to create say, x threads, each with their
own instance of the interpreter?

what is the PyDict format of globals/locals from PyModule_GetDict?
what vars are declared in this ModuleDict? is it necessary to add
anything there?

does PyCodeObject keep (or represent) the compiled code in mem?
would it be worth maintaining a heap of PyCodeObjects (one for
each .pyx, or until cache is full...) as opposed to recompiling
for each request (unless it is desired, say, .pyd), say
PyCodeObject  * pcoCached[] = malloc( 500000 );
will Py use this new heap for the CodeObject pointers aswell as?
will the CodeObjects fill the heap, or will Py, when compiling
for a CodeObject, create mem elsewhere? eer, does sizeof(PyCodeObject)
include the Py bytecode?

if i were to maintain the PyCodeObjects, how then would i have
an instance of the interpreter run this code? ahh, obviously
PyEval_EvalCode, but what if Python is handling the sub-interpreters?
how to say, ThatSubInterpreter->EvalCode( PyCodeObject )?

is Py_NewInterpreter really a 'fresh copy' of the main interpreter
for use by a thread? if i add built-in mods with PyImport_AddModule,
do i need to do this for each Py_NewInterpreter? what about
Py_InitModule? does this need to be called for each new aswell?

in the examples, the telnet server seems to call Py_NewInterpreter
for each request. would i have to, if say i wished to limit the
worker threads, when a request occurs, if threads >= max, put
request in queue, OR, create a new thread with Py_NewInterpreter,
handle the request, check to see if any new requests are in the
queue [and handle them too], then Py_EndInterpreter? what about
reusing Py_NewInterpreters after they've been idle for a while?
(say, a LIFO q of interpreters... you know how it is... no action
for a while then bang, everybody wants in on your server!!)

cheers and thanks in advance
bryn





From mal at lemburg.com  Fri Aug 13 12:26:24 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 13 Aug 1999 18:26:24 +0200
Subject: L postfix char on longs - what is the best way to discard it.?
References: <934546409.21489.0.nnrp-14.d4e5b0e4@news.demon.co.uk>
Message-ID: <37B44730.4C65E3E9@lemburg.com>

Jim Crumpler wrote:
> 
> Hi folks - I'm batting to find a nice clean way to remove the 'L' from long
> repr strings .. (eg, my SQL queries end up like "SELECT * FROM blah WHERE
> guff = 23423L"..)
> 
> I can fiddle with every bit of my code that attempts to display a long, such
> as
> 
> query = "... WHERE guff = %s" % repr(long(x))[0:-1]
> 
>  -- its ugly when you use alot of longs, but does the job..
> 
> I can bugger around with long_format() to remove the postix 'L' altogether,
> or build a % substitution to print a long without the L  - uglier for
> distribution, but nice for writing code - I have no idea what repercussions
> this has on the library code - probably bad.. I'm assuiming the actual
> reason for the 'L' is so a string could be then fed back as a literal
> without overflow the default integer conversion.  Couldn't the integer
> conversion build a long on overflow, or would that get uglier?
> 
> Does anyone have some suggestions on better methods?

You could use mxODBC which takes care of all this for you: just
pass in the integer or long object as variable, e.g.
cursor.execute('SELECT ... WHERE id = ?',(x,)).

If the id column is a BIGINT column, x is converted to a string before
passing it to the database. For long integer objects, the trailing 'L'
is removed automagically. In case id is a normal INT column,
the equivalent of int(x) is passed to the database as C long integer.
Overflows raise an exception. Note that SQL BIGINT columns usually
only support 64bit integer values.

mxODBC can be found on my Python Pages.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   140 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From stephan at pcrm.win.tue.nl  Tue Aug 31 11:25:40 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 31 Aug 1999 17:25:40 +0200
Subject: problem with execfile
References: <37CBE760.3227B3A0@GermanLloyd.org>
Message-ID: <m3g11055vv.fsf@pcrm.win.tue.nl>

Berthold Hoellmann <hoel at GermanLloyd.org> writes:

> Hello,
> 
> I have a problem with execfile:
> 
> python
> Python 1.5.2 (#2, Jul  7 1999, 09:18:40) [C] on sunos5
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> def x():
> ...     execfile("ZustandsDaten.dat")
> ...     print dir()
> ...     print Tiefgang
> ... 
> >>> x()
> ['Pitch', 'Tiefgang', 'U', '__doc__', 'array', 'n']
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "<stdin>", line 4, in x
> NameError: Tiefgang
> >>> 
> 
> Why is Tiefgang unknown in the function x?

This one is very subtle. Here's my take on it.
(But I'm not Tim, so I might be wrong.)

The problem is in a compile-time optimization made by
Python. Any local variables (like Tiefgang) are put
in a special array that's much faster than the ordinary
locals() dictionary. But execfile() uses the ordinary locals()
dictionary. So when "print Tiefgang" happens, Python tries
to look it up in the special array. But it's not there,
so a NameError is generated.

Is this a bug or a feature? I don't know. But I do know
how to kludge around it: just add a phony "exec" statement,
something like `exec "1"'. This will prevent the compiler
from doing this optimization, and then your code will work.

Greetings,

Stephan






From chinaski_py at my-deja.com  Sat Aug 21 06:58:14 1999
From: chinaski_py at my-deja.com (chinaski_py at my-deja.com)
Date: Sat, 21 Aug 1999 10:58:14 GMT
Subject: NEWBIE:Let's try that again..Help w/path
Message-ID: <7pm0o4$e0h$1@nnrp1.deja.com>

I'm new and need help with my Python path.When I go to open a Python
file (.py) it won't open .I go to (on Win98):
start
run
sysedit
and here is my path:
path %path%;C:\Perl\bin;path=C:\PROGRA~1\Python\;%path%
                Any help on this would be appreciated,
                                G.Norton (chinaski_py)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From BC at prism.co.nz  Tue Aug 24 05:24:18 1999
From: BC at prism.co.nz (Bing Chen)
Date: Tue, 24 Aug 1999 17:24:18 +0800
Subject: newbie: Why Py_Initialize( ) doesn't work in VC6.0
Message-ID: <7ptah7$hvl$1@newsource.ihug.co.nz>

Hi, all
I try to emed Python into my application, follow the manual setting the
project in VC6.0. In Object/library modules: python15.lib
                              in Additional include directories: d:\program
files\python\include
                             In Main program: #include "python.h"
                                                           Py_Initialize()
When I debug it, always show me "CXX0017:Error:symbol 'Py_Initialize' not
found.
What's happen?

Many thanks!
Bing





From skip at mojam.com  Tue Aug 24 14:43:55 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 24 Aug 1999 13:43:55 -0500 (CDT)
Subject: Python equivilent of pack()?
In-Reply-To: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com>
References: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com>
Message-ID: <14274.59264.706875.200701@dolphin.mojam.com>

    Bit> My question is this: is there a functional equivilent to the Perl
    Bit> pack() and unpack() functions in Python? Or even better, is there
    Bit> an easier way to do what I want with Python? If not, how many
    Bit> people think this would be a useful addition to the Python standard
    Bit> library?

Bit,

Check out the struct module docs at:

    http://www.python.org/doc/lib/module-struct.html

It'll most likely do just what you want.  It even calls its methods "pack"
and "unpack", so it should be fairly familiar to your Perl neurons...

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...




From donn at u.washington.edu  Fri Aug 27 13:40:50 1999
From: donn at u.washington.edu (Donn Cave)
Date: 27 Aug 1999 17:40:50 GMT
Subject: Why do we call python a scripting language?
References: <37c6bb11.340373891@midpoint> <37c66347.518331@news.netset.com>
Message-ID: <7q6ij2$vm4$1@nntp6.u.washington.edu>

Quoth tdixon.no at spam.fwi.com (Timothy Dixon):
| On Fri, 27 Aug 1999 10:18:55 GMT, guy_oliver at yahoo.com wrote:
|> Why do we call python a scripting language?  
|>
|> For me, and for most of the people I try to tell about the benefits of
|> python, the title 'scripting language' implies that its not a real
|> language capable of real programs.  It implies to most of them that it
|> is good for simple sysadmin and maybe short cgi scripts.
|
| I put it in the class of scripting languages because it's embeddable
| (as I understand it; I'm actually just a newbie) in other applications
| where it can be used for scripting another app.

Yes!  This is the useful meaning for "scripting", as I see it!  The
definitions that try to distinguish between ``quick, inconsequential''
vs. ``serious, production'' etc. are hopelessly lame, regardless of
their distinguished sources, because it's a useless struggle to assign
importance this way.  A 3 line Bourne shell script can be mighty
important production software.

I think this fits in with the other contexts where people use the term,
too.  I'm mostly a UNIX geek, but on platforms like MacOS, BeOS, AmigaDOS,
(Windows?) etc. I gather ``scripting'' is very much tied to the context
of some application.

But in these contexts the scripting language isn't always embedded, there
may (should) be other ways for it to be integrated with its application.
And as an embedded application language frankly I think Python isn't really
at its best.  The plasticity of languages like Lisp and Tcl make them easier
to adapt in some ways.  Like the UNIX shell is adapted to its application
environment, UNIX commands.

| It's not *just* in that class, of course; polymorphism and multiple
| inheritance are great real-world characteristics. :-)

Fer sure.  My vote would be to leave ``scripting'' out of it, or use
the term only with explanation.  The references to rapid prototyping,
etc., are somewhat along the lines I think people really have in mind
here, and perhaps someone who's fond of the term can articulate a useful
meaning for it without making it sound unduly trivial.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From greg.ewing at compaq.com  Tue Aug 31 20:56:37 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 01 Sep 1999 12:56:37 +1200
Subject: Returning none
References: <000001bef265$fe64f340$192d153f@tim>
Message-ID: <37CC79C5.1BFDC710@compaq.com>

Tim Peters wrote:
> 
> You could invent a new internal NotARealObject type and
> return a pointer to an object of that type, but then it gets messier.

Not if you go about things the right way!

The compiler knows whether the result of any given
call gets used or not. Suppose it generates CALL_FUNCTION
when the result is used, and a new opcode CALL_PROCEDURE
when it doesn't.

CALL_PROCEDURE checks that the result is NotARealObject 
(I'll call it Void from now on) and raises an exception 
otherwise. CALL_FUNCTION does the opposite.

> is it also OK for f() to return nothing in the statement
> 
>     f(), g(), h()
> 
> ?  You must allow that else break existing code,

We've *already* broken existing code, since we're now 
disallowing any unadorned expression whose value isn't 
Void. 

Strictly speaking, in that example f() would not be
allowed to return Void, since it's being called as a
function. But, that whole statement could be statically 
rejected by the compiler, since it always evaluates to
a tuple.

Moreover, the compiler could (and would almost have to)
reject any unadorned expression that wasn't a call
or something trivially equivalent (such as a call
surrounded by parentheses).

In summary, a Void object *could* have been incorporated
into the scheme of things and made to Do The Right Thing.
But then it would have been a different language.

Maybe in Python 2?

Greg



From kosullivan at snet.wit.ie  Tue Aug  3 07:53:58 1999
From: kosullivan at snet.wit.ie (Kieran O' Sullivan)
Date: Tue, 3 Aug 1999 12:53:58 +0100 (IST)
Subject: Python BSCW Server Error AttributeError: altzone 
Message-ID: <Pine.LNX.4.10.9908031250400.31183-100000@snet.wit.ie>

I am running BSCW on SuSE Linux 5.3 with Apache 1.3.  The BSCW will
allow logins and creation of users but when ever a user whether they 
are an Admin or just na ordinary user tries to create a folder the
following happens.
 
Unanticipated Error (BSCW Release 990623-1553):
Traceback (innermost last):
  File "wrapped", line 16, in ?
    cl_request.run('bs_cgi')
  File "BSCW3/pyc15/cl_request.py", line 1351, in run
  File "BSCW3/pyc15/cl_request.py", line 801, in handle_rq
  File "BSCW3/pyc15/cl_request.py", line 891, in handle_it
  File "BSCW3/pyc15/op_get.py", line 65, in handle_GET
  File "BSCW3/pyc15/cl_core.py", line 1166, in get
  File "BSCW3/pyc15/bs_container.py", line 150, in get
  File "BSCW3/pyc15/bs_container.py", line 274, in visible_entries
  File "BSCW3/pyc15/bs_container.py", line 274, in <lambda>
  File "BSCW3/pyc15/cl_core.py", line 98, in get_as_entry
  File "BSCW3/pyc15/cl_core.py", line 946, in get_as_entry
  File "BSCW3/pyc15/bs_artifact.py", line 261, in get_as_entry
  File "BSCW3/pyc15/bs_artifact.py", line 91, in get_entry_params
  File "BSCW3/pyc15/bs_interface.py", line 438, in event_msg
  File "BSCW3/pyc15/bs_utils.py", line 304, in relative_time
  File "BSCW3/pyc15/cl_request.py", line 295, in localtime
  File "BSCW3/pyc15/cl_request.py", line 914, in __getattr__
  File "BSCW3/pyc15/cl_request.py", line 1200, in def_timezone

AttributeError: altzone

Once this has happens that user is no longer able to log into the server.

Whne you try to edit details of the user, as an admin user you get

Unanticipated Error (BSCW Release 990623-1553):
Traceback (innermost last):
  File "wrapped", line 16, in ?
    cl_request.run('bs_cgi')
  File "BSCW3/pyc15/cl_request.py", line 1351, in run
  File "BSCW3/pyc15/cl_request.py", line 801, in handle_rq
  File "BSCW3/pyc15/cl_request.py", line 891, in handle_it
  File "BSCW3/pyc15/op_chdetails.py", line 118, in handle_GET
  File "BSCW3/pyc15/cl_request.py", line 914, in __getattr__
  File "BSCW3/pyc15/cl_request.py", line 1188, in def_localtime_now
  File "BSCW3/pyc15/cl_request.py", line 295, in localtime
  File "BSCW3/pyc15/cl_request.py", line 914, in __getattr__
  File "BSCW3/pyc15/cl_request.py", line 1200, in def_timezone
AttributeError: altzone

A user account may be deleted but then if you try to recreate that user
with the same email address the server won't accept the address.

Since the BSCW server was written in python I thaught that some one on
this list might have experience of this problem. 
_________________________________________________________________
Visit my website at http://snet.wit.ie/kieran





From philh at vision25.demon.co.uk  Tue Aug 10 18:44:25 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 10 Aug 99 22:44:25 GMT
Subject: Python GUI
References: <19990805141904.115.qmail@hotmail.com> <37AB0600.869DFD87@NOSPAM.strs.co.uk>
Message-ID: <934325065snz@vision25.demon.co.uk>

In article <37AB0600.869DFD87 at NOSPAM.strs.co.uk>
           I.Clarke at NOSPAM.strs.co.uk "Ian Clarke" writes:
> > > > Is there a Python GUI builder out there anywhere?
> > >Not that I'm aware of.
> > >Anyone know of one in-the-works?
> > http://www.pythonware.com/products/works/
> 
> Unfortunately it doesn't seem to be open source :-(
> 
> I have often thought that a Python based IDE would be an ideal
> OpenSource answer to Visual Basic.

I wholeheartedly agree; Python is the best language around, for most
tasks that aren't too computationally intensive.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From skaller at maxtal.com.au  Wed Aug  4 02:24:59 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Wed, 04 Aug 1999 06:24:59 GMT
Subject: 1.5.2 for: else:
References: <1279007788-11957995@hypernet.com> <379F6D43.3531FE0@compaq.com>
Message-ID: <37a7dc67.1803766539@news.triode.net.au>

On Thu, 29 Jul 1999 08:51:15 +1200, Greg Ewing <greg.ewing at compaq.com> wrote:

>Gordon McMillan wrote:
>> 
>> But finding a match in a list, and testing whether you
>> fell off the end without finding one, is (without the else clause) a
>> much messier proposal.
>
>Not if you go about it the right way. I always
>put such loops in a procedure of their own,
>and use return:
>
>  def find_the_holy_grail(stuff):
>    for x in stuff:
>      if is_holy_grail(x):
>        return x
>    return None

Unfortunately, python lacks lexical scoping, which means this technique
cannot be easily generalised. In particular, if the loop body requires
access to the current context, you can't use a function to encapsulate it.


John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From aahz at netcom.com  Fri Aug 20 10:40:10 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 20 Aug 1999 14:40:10 GMT
Subject: PIGgies?
References: <19990818171523.A22207@digex.net>
Message-ID: <7pjpca$1a5@dfw-ixnews15.ix.netcom.com>

In article <19990818171523.A22207 at digex.net>,
John Landahl  <jpl at digex.net.nospamplease> wrote:
>
>Lately I've been considering forming a high profile user group in San
>Francisco to help promote Python as a serious tool for mainstream use.
>Much to my chagrin I discovered that there's already a (recent) group
>called "BayPIGgies".  I later noticed that the term "PIGgies" is used
>elsewhere for Python user groups.  Am I the only Python enthusiast that
>has a problem with the term "PIGgies"?

I suspect that you've fallen for the common fallacy of thinking that
"Python" refers to a species of snake; in fact, "Python" is a reference
to Monty Python's Flying Circus, and most pythonistas make a point of
heavily using irreverent humor.

There's nothing like a cream pie to the face for demonstrating how
serious you are.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From cgw at fnal.gov  Sat Aug 21 15:53:29 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Sat, 21 Aug 1999 19:53:29 GMT
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <000301beebac$792199a0$0c2d2399@tim>
References: <000301beebac$792199a0$0c2d2399@tim>
Message-ID: <14271.953.466287.938800@buffalo.fnal.gov>

 > [Christopher Browne]
 > > Is it so much work to come up with a half-reasonable card shuffling
 > > algorithm?

Apparently so, since so few of the solutions posted (including yours)
have biases:

 > >
 > > The one I've always seen cited as providing decent behaviour looks
 > > thus:
 > >
 > > for (i = 0; i > n; i++) {  /* surely intended "i < n" */
 > >    swap (item[i], item[random(n)]);
 > > }
 > 

I fed this to my statistics-tester.  You can see that you tend to
favor odd permutations when the list is of an odd length and even
permutations when the list is of an even length.  You're swapping too
much, in essence.  The bias in the sign of the generated permutations
becomes more marked as the list becomes longer.

Also, the variances seem to be higher for elements later in the list,
although I'm not sure if the results below are statisticall
significant.

What these numbers are is the result of scrambling the sequences
[0,1], [0,1,2], ..., [0,..7]
each 50000 times, and recording the signs (+1/-1) of the generated
permutations and the means and RMS variances of each sequence element.
See my previous posts in this thread for more...

Note also that if you have a permutation algorithm that tends to favor 
elements from a subgroup (e.g. even permutations) then repeated
applications of the algorithm won't help you at all.



#Chris Brown's method
def brown(seq):
    for i in range(len(seq)):
        j=random.randrange(0,len(seq))
        seq[i],seq[j]=seq[j],seq[i]

Testing <function brown at 81604f0>
len= 2
 mean values= [0.49874, 0.50126]
 rms= [0.499998412397, 0.499998412397]
 mean sign= 0.00252
len= 3
 mean values= [0.9641, 1.03796, 0.99794]
 rms= [0.792698675412, 0.839523101767, 0.81492070559]
 mean sign= -0.03528
len= 4
 mean values= [1.41586, 1.52312, 1.55922, 1.5018]
 rms= [1.07712601881, 1.11809904105, 1.15433660238, 1.11626016681]
 mean sign= 0.06644
len= 5
 mean values= [1.86166, 1.99346, 2.06626, 2.08022, 1.9984]
 rms= [1.36447867129, 1.38496831314, 1.43166672532, 1.46173347489,
1.41552726572]
 mean sign= -0.07372
len= 6
 mean values= [2.31702, 2.43602, 2.56682, 2.59322, 2.58368, 2.50324]
 rms= [1.65208907738, 1.65539921457, 1.70093359294, 1.74600974556,
1.76338244791, 1.70917216874]
 mean sign= 0.0886
len= 7
 mean values= [2.77782, 2.91118, 3.0239, 3.0912, 3.10874, 3.09468,
2.99248]
 rms= [1.94859335101, 1.92894038467, 1.9772022633, 2.01373348783,
2.05070612531, 2.06344268096, 1.99186933547]
 mean sign= -0.09652
len= 8
 mean values= [3.2336, 3.37306, 3.46918, 3.57226, 3.6168, 3.63026,
3.6131, 3.49174]
 rms= [2.23409736583, 2.21326144782, 2.23776900676, 2.28316852037,
2.32304062814, 2.3567503755, 2.35576068182, 2.2918751651]
 mean sign= 0.10684



From jeffp at crusoe.net  Wed Aug 11 12:56:28 1999
From: jeffp at crusoe.net (Jeff Pinyan)
Date: Wed, 11 Aug 1999 12:56:28 -0400
Subject: os.environ => invalid token
In-Reply-To: <7os33a$147t@enews4.newsguy.com>
References: <7os33a$147t@enews4.newsguy.com>
Message-ID: <Pine.GSO.4.10.9908111253020.15790-100000@crusoe.crusoe.net>

[posted & mailed]

On Aug 11, Peter Torstenson blah blah blah:
> I tried the os.environ['HOME'] statement from the libref. man to get a
> string with the home directory.

Are you sure you imported the os module?

  >>> import os
  >>> print os.environ['HOME']
  /home/jeffp

Works fine for me.  What OS are you on?

You can also check what python sees as your environment by doing:

  >>> import os
  >>> for n in os.environ.keys():
  ...   print n + "=" + os.environ[n]

--
jeff pinyan    japhy at pobox.com  japhy+perl at pobox.com  japhy+crap at pobox.com
japhy's little hole in the (fire) wall:       http://www.pobox.com/~japhy/
japhy's perl supposit^Wrepository:       http://www.pobox.com/~japhy/perl/
The "CRAP" Project:                 http://www.pobox.com/~japhy/perl/crap/
CPAN ID: PINYAN           http://www.perl.com/CPAN/authors/id/P/PI/PINYAN/




From jimn at minusen.force9.co.uk  Tue Aug 10 16:40:42 1999
From: jimn at minusen.force9.co.uk (Jim Nicholson)
Date: 10 Aug 1999 21:40:42 +0100
Subject: internet programming with python
References: <37AF678E.CC8E165B@netscape.com>
Message-ID: <m2vhanv06t.fsf@minusen.force9.co.uk>

Vijay Mallela <vijaym at netscape.com> writes:

> Hi,
> I am looking for the book "Internet programming with python". I need it
> at the earliest..
> I checked out amazon and other stores and they dont have it.. does any
> one know where I can get the book

it would help to know whereabouts in the world you are!  If I saw a
copy in a bookshop, it's not much use telling you if you're the other
side of the world!

Jim



From skip at mojam.com  Tue Aug 17 10:43:37 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 17 Aug 1999 09:43:37 -0500 (CDT)
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <7pb1eb$elt$1@nnrp1.deja.com>
References: <7pb1eb$elt$1@nnrp1.deja.com>
Message-ID: <14265.29899.470021.309355@dolphin.mojam.com>

    Ole> I've been writing an algorithm myself, but it's incredibly slow and
    Ole> I would like something quicker (of course :) )

Got this off the list a couple years ago.  Does the trick for me:

    def shuffle(lst):
	"randomize a list"
	for i in range(len(lst)-1,0,-1):
	    choice = int(whrandom.random() * i)
	    lst[choice], lst[i] = lst[i], lst[choice]

    >>> l = range(20)
    >>> shuffle(l)
    >>> l
    [4, 17, 12, 6, 7, 15, 1, 11, 16, 10, 5, 3, 9, 19, 0, 8, 18, 13, 14, 2]

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From cbbrowne at news.hex.net  Fri Aug 20 23:49:57 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Sat, 21 Aug 1999 03:49:57 GMT
Subject: Unsorting(randomizing) a sequence
References: <000701bee917$d8161cc0$eaa22299@tim> <14268.10663.137315.893101@buffalo.fnal.gov>
Message-ID: <Flpv3.16185$gO1.546412@news2.giganews.com>

Is it so much work to come up with a half-reasonable card shuffling
algorithm?

The one I've always seen cited as providing decent behaviour looks
thus:

for (i = 0; i > n; i++) {
   swap (item[i], item[random(n)]);
}

This algorithm guarantees that each and every item on the list gets
swapped with something.

If I wanted to make sure the permutation *really* looks random, I'd
consider running this "shuffle" multiple times, and would make sure I
had a RNG of decent quality with a period Rather Longer than n.
-- 
"For systems, the analogue of a face-lift is to add to the control
graph an edge that creates a cycle, not just an additional node."
-- Alan Perlis
cbbrowne at hex.net- <http://www.hex.net/~cbbrowne/languages.html>



From kj0 at mailcity.com  Wed Aug  4 13:23:34 1999
From: kj0 at mailcity.com (kj0)
Date: 4 Aug 1999 17:23:34 GMT
Subject: "hello, world" of embedding/extending?
Message-ID: <7o9sum$2ak$1@news.panix.com>




What are the "hello, world" equivalents of embedding and extending
Python (in/with C or C++)?

More generally, where can I find source code didactic examples of
embedding and extending Python?  (I write "didactic" because I'm not
quite at the level yet of being able to read real-world,
multiplatform, distribution-grade applications; rather, I'm interested
in examples ranging from the "hello, world" level of complexity to the
level of a full-blown useful application that someone may write for
personal use, i.e. not intended for wide distribution).

Thanks!

KJ, recovering perloholic




From ivnowa at hvision.nl  Mon Aug  2 04:22:12 1999
From: ivnowa at hvision.nl (Hans Nowak)
Date: Mon, 2 Aug 1999 08:22:12 GMT
Subject: Call for snippets
Message-ID: <199908020821.KAA07953@axil.hvision.nl>

Hello,

Some of you may have noticed that I have a snippets site since a few 
months. (In this definition, snippets are pieces of working Python 
code, usually demonstrating something or solving a problem.) Until 
now, I have extracted them from newsgroup messages. This isn't an 
ideal situation, because there are lots of messages which I all have 
to read, and I don't really have the time for that. (Yeah I know... 
sacrilege... :^)

This isn't really a problem though. What does bother me is that my 
collection isn't growing very rapidly. I have 119 snippets online 
now, with 100+ still residing on my harddisk, untested. It's a start, 
but there really isn't much progress, mainly because of the reasons 
mentioned above. I have a feeling that there is a lot of good and 
useful code there but that I have to wait until it accidentally shows 
up in the newsgroup, triggered by somebody's question.

So, my request is obvious. I would ask everyone to browse through 
their code and see if they have any interesting, useful or just plain 
cool snippets. I'm not really picky; a nifty function would do. I am 
not really looking for complete modules since I have a snippet 
repository, not a module repository (nor anything like CPAN, the 
pyPit, etc), but I won't refuse them. In fact, anything that solves 
any problem could be a valuable snippets. I'd almost say, 'when in 
doubt, post your code'. <g>  I'd go even farther: if you have Python 
code somewhere and think there might be interesting functions in it, 
I would be pleased if you pointed me to it; I will scan and extract 
them myself.

My goal is to have a site where both newbies and more advanced coders 
could go to see how to do something in Python, or how to do something 
*differently* if they already have some solution, because more 
snippets can tackle the same problem. But in order to do that I need 
some more code! Or at least a larger base. <wish> Now if everyone 
reading this ng would post at least one snippet, there would be an 
explosive growth... </wish>

?ou can check out the page at http://www.hvision.nl/~ivnowa/snippets/ 
. It's (of course) still under construction. Snippets and comments 
can be sent to (feeble spam protection) ivnowa @ hvision . nl or 
hnowak @ cuci . nl. (The address mentioned on the side is actually 
obsolete. I will remove it.)

Thanks in advance,

Hans Nowak (ivnowa at hvision.nl)
Homepage: http://fly.to/zephyrfalcon



From junkster at rochester.rr.com  Tue Aug 24 11:01:16 1999
From: junkster at rochester.rr.com (junkster at rochester.rr.com)
Date: Tue, 24 Aug 1999 15:01:16 GMT
Subject: Question, re: Python & Windows NT
Message-ID: <7puc3o$29d$1@nnrp1.deja.com>

Hi!

	I want to check something out, but I haven't found anything
in the docs of python to verify this....

	Will SHUTIL's COPYFILE command copy extended attribute
information under Windows NT?

	What I'm trying to do is modify FTPMIRROR to COPYFILE instead, which
should _not_ be difficult.  But the information will be copied from a
MACFILE VOLUME from a NT SERVER, which means that it should have extended
attribute information (creator type, etc).

	If COPYFILE won't copy that data, then I might as well, stick with
FTPMIRROR and save myself the trouble.

	Keep in mind, both machines ("server" & "Client") will both be
Windows NT Server's (v4), with the same version of python (1.52).

	If copyfile won't do this, then does anyone have a suggestion on how
to do this?  I'm trying to mirror the server, and DIRECTORY REPLICATION
(from NT) isn't robust enough to deal with all the data.

			- Benjamin
			Junkster at rochester.rr.com


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From ray at unival.com  Tue Aug 17 20:08:23 1999
From: ray at unival.com (ray at unival.com)
Date: Tue, 17 Aug 1999 17:08:23 -0700
Subject: Announcement: Portland Oregon user group meeting
Message-ID: <882567D1.00007994.00@bilbo.collegenet.com>

The first meeting of the Portland, Oregon area Python interest group
(PORPIGgies?) will be held Tuesday September 7, 1999 from 7:00pm to 9:00pm
in the Benjamin Franklin Building, downtown Portland.

The inaugural speaker will be Amos Latteier of Digital Creations speaking
on "Zope and XML".  He'll discuss the  Zope application server and it's web
object system from the perspective of XML.
  * DOM interface to standard Zope objects.
  * Using XSL as an object query language.
  * The XML Document Zope product, and how it exposes XML data to Zope.
  * Cataloging XML data.
  * Where XML and Zope are going.
The discussion will cover some basics about Zope's object system, but will
focus on its integration with XML. Even if you don't know Zope, if you have
some idea of what XML is, you should be OK.

The initial goals of the group are to meet the first Tuesday of each month,
and have a fairly technically oriented meeting. The meeting format will
consist of one or two technical presentations each followed by Q&A.
Socialization will be encouraged after the presentations at a local pub.
Admission is free.

The entrance to the Benjamin Franklin Building is located on the corner of
SW 1st Ave and SW Columbia Street in Portland. The building entrance is
monitored by a security guard so it is very important that you RSVP so you
can be listed at the security desk.  I'll be in the lobby to greet visitors
before 7:00, after that the guard will check the entrance every 5 to 10
minutes.  So, please try to be here before 7:00  The address of the
building is 1 SW Columbia Ave, Portland, OR 97258

Street parking is free after 6:00pm, or there is a parking structure at SW
1st and Jefferson.

Please RSVP by email to Ray at unival.com

Ray Price







From skip at mojam.com  Tue Aug 31 19:52:44 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 31 Aug 1999 18:52:44 -0500 (CDT)
Subject: Parrot-0.0.1
In-Reply-To: <7qh7b6$g9g$1@wagner.wagner.home>
References: <935279776snz@vision25.demon.co.uk>
	<7ps4ma$n06$1@wagner.wagner.home>
	<935457373snz@vision25.demon.co.uk>
	<7qh7b6$g9g$1@wagner.wagner.home>
Message-ID: <14284.26956.792223.975584@dolphin.mojam.com>

    Victor> But in tcl you can do

    Victor> option add mywindow*Button.Background red and achiveve the same
    Victor> effect. Moreover, in Parrot it seems to be coded into app, but
    Victor> with Tcl option command it can be overriden by user Xdefaults
    Victor> file.

Don't forget Parrot (and tcl/tk) will run on other systems besides X.  X
defaults files are not very platform-independent.  I suspect Parrot will
have to rely on something other to allow users to override defaults.
Perhaps it will map onto an X defaults file on X/Unix systems, but will
probably map to something else on Windows and Mac.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From philh at vision25.demon.co.uk  Mon Aug 30 13:36:09 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 30 Aug 99 17:36:09 GMT
Subject: Parrot: Why the clunky syntax?
References: <935279776snz@vision25.demon.co.uk> <935454399snz@vision25.demon.co.uk> <37C2B9CE.95AA5FEC@lri.fr> <37C9FF53.DD466783@compaq.com>
Message-ID: <936034569snz@vision25.demon.co.uk>

In article <37C9FF53.DD466783 at compaq.com>
           greg.ewing at compaq.com "Greg Ewing" writes:
> Instead of a syntax like
> 
>   window @MyWindow "My First Window" {
>        menuBar {
>           menu "File" {
>              menuItem @New "New"
>           }
>   }
> 
> why not something more Python-inspired, e.g.
> 
>   window MyWindow("My First Window"):
>     menubar:
>       menu File:
>         menuitem New
>
> After all, we like Python because it does without
> line noise like @{} -- don't we?

That's a very good question.

Basically, the current syntax was chosen because I like it. I don't
think there would be much problem implementing a more Pythonesque
syntax -- probably just rewriting the lexical analyser so it
output INDENT and DEDENT symbols would do it. In any case, Parrot is
explicitly designed to be able to cope with multiple frontends and 
backends; perhaps one could be added.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From gmcm at hypernet.com  Fri Aug  6 16:51:15 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 6 Aug 1999 15:51:15 -0500
Subject: Help! Getting used memory
In-Reply-To: <7ofdbj$m6u$1@nnrp1.deja.com>
Message-ID: <1278153615-27035363@hypernet.com>

Sunit writes:

>       I have a 'renegade' process which I think doesn't release
>       memory
> after it quits. Is there a way to check this user win32 functions.
> If so, could someone please let me know how to do this in python.
> I'm guessing I might have to use calldll...

When a process ends, all memory owned by that process is guaranteed 
to be released. There aren't that many ways to leak after process 
death. You could be leaving junk on the clipboard, or have started 
COM stuff and screwed up the refcounting so it stays around. On Win9x 
I suppose you could screw up the 16 bit subsystems. You certainly 
can't leak malloc'ed memory past process death.

- Gordon



From jstevens at basho.fc.hp.com  Fri Aug 13 23:07:27 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:07:27 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132016.OAA03573@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:16 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cf5f@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :But, once again, not only is this not a valid criticism but in
> :point of fact Python is more correct than Perl for this
> :example.
> 
> It is wrong from the Larry's principle that things that behave
> differently should have different appearances.

Then why doesn't Perl have two different operations for
assign/reassign, vs. add to an array?  A tiny little violation of
Larry's principles, that.

John S.




From philh at vision25.demon.co.uk  Thu Aug 26 16:30:30 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Thu, 26 Aug 99 20:30:30 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net> <37C552DF.9F90BA36@megsinet.net> <37c563bc.26497369@news.hydro.co.uk>
Message-ID: <935699430snz@vision25.demon.co.uk>

In article <37c563bc.26497369 at news.hydro.co.uk>
           gary at onegoodidea.com "Gary O'Keefe" writes:
> A keyboard was whacked upside David Oppenheimer's head and out came:
> 
> >I don't quite understand the relevance of this entire Perl advocacy string
> >here.  If Xah thinks that getting mean responses to postings (especially
> >postings from newbies) is so very cool, then why waste time over here doing
> >postings to the Python newsgroup.  I must say that in most newsgroups I have
> >found people friendly and helpful.  If I were to post to a newsgroup and get
> >flamed for not knowing anything or given obviously incorrect information just
> >to piss me off, I'd never go back.  It would not generate feelings of
> >advocacy for me.  Who the hell wants to have some intellectual elitist
> >assholes tell you you're ignorant just for asking a question?!
> 
> Python is a language that you hear about after conversing with
> knowledgeable people, a language recommended by experts to the keen
> computer scientist -- I've always thought of Python, perhaps unfairly,
> as being the language for those people interested in language design
> as an end in itself. But, one way or another, this makes for edifying
> conversation between educated people.
> 
> Perl on the other hand is the first language anyone even vaguely
> interested in putting together an interactive web-server hears about.
> That means EVERY DUMBASS IN THE UNIVERSE who wants to put their cock
> on the internet and have you virtually stroke it but, importantly,
> can't be bothered to actually learn all this computer bullshit posts
> to comp.lang.perl.misc.
> 
> If your newsgroup (that you'd just paid to download) was filled with
> posts like this:
> 
> > Subject: [newbee] I am too damn lazy - can't you do this for me?
> >
> > hay gurus
> >
> > Im starting up an intractive cock-stroking website
> > (http://www.cockstroke.com/). Its going 2 B KEWL!!!
> >
> > But I NEED A CGI TO MAKE IT WORK!!!
> >
> > Can you supr-clever gurus write me something??!! It woudnt take
> > you any time and ITD B GRATE!!!
> >
> > THX
> > 
> > Newbee
> 
> then your patience would vapourise, just like ours.

Hey, I thought all Perl users were like that :-)

> Gary
> 
> The newbee post above and any mentioned websites are purely fictional.
> Any similarity to a real newbee, (living or dead) or website are
> purely co-incidental.
> 
> <NOT SERIOUS>
> p.s. I've got a great idea: let's teach those smug python bastards a
> lesson by re-directing all the newbees to comp.lang.python. Tell them
> it's the latest, greatest, thing for developing their CGI scripts.

I can see it now: Python, the language for a firmer, harder, hotter,
cock-stroking experience!!!

-- 
Phil Hunt....philh at vision25.demon.co.uk




From bkc at Murkworks.com  Thu Aug  5 11:00:26 1999
From: bkc at Murkworks.com (Brad Clements)
Date: Thu, 5 Aug 1999 11:00:26 -0400
Subject: Python on a Novell Netware, Netscape web server
References: <mt2.0-15236-933815857@news.informatik.uni-bonn.de>
Message-ID: <7oc8v7$bii$1@news.clarkson.edu>

I am working on a port of Python to NetWare as an NLM. Its still very crude.
I have almost all of the NDS api swig'd for both client-side development and
python.nlm

However the best solution to this is to use UCS (Universal Component
Software) api to interconnect Python with Netbasic, Perl, Netscape and other
services (and vice-versa). I'm still plying Novell to get the necessary
APIs.

Right now this is a NW5 solution (NW4 doesn't support virtual memory, which
could be a problem for Python depending on your application).

Also, I don't have modules working as .lib NLMs yet, so everything must be
precompiled into one giant size python.nlm

Without module finalization, I can't free memory when Python is unloaded,
hence netware complains about unreleased resources.

--
Brad Clements,
bkc at murkworks.com
Bret Orrell <breto at waymark.net> wrote in message
news:mt2.0-15236-933815857 at news.informatik.uni-bonn.de...
> I have a Novell Netware 4.11 sever  at work running Netscape Enterprise
> server 3.5.
>
> The Netscape server comes with Perl and there is a CGI2PERL.NLM module
that
> loads when it needs to run a Perl script.  Is there a version of Python
for
> this?
>
> Python.org lists just about every platform except for Novell and Netscape.
>
> I'm not real fimilar with setting up a web server but this project got
> dumped on me anyways.





From sjmachin at lexicon.net  Fri Aug 27 19:18:47 1999
From: sjmachin at lexicon.net (John Machin)
Date: Fri, 27 Aug 1999 23:18:47 GMT
Subject: Fuzzy string matching?
In-Reply-To: <012501bef061$10db8b30$f29b12c2@secret.pythonware.com>
References: <012501bef061$10db8b30$f29b12c2@secret.pythonware.com>
Message-ID: <19990827231353095.AAB161.428@max41051.izone.net.au>

On 27 Aug 99, at 9:52, Fredrik Lundh wrote: [[pruned]]
> Hans Nowak <hnowak at cuci.nl> wrote: [[pruned]]
>> I expect that some of their input will not
> > be recognized in the card database. In such cases, I would like to
> > present the user with a list of card names which match the input
> > closely, but not entirely... strings that 'almost match'. OK, so it's
> > just a fuzzy string match. Is such a thing done in Python already? If
> > not, would it be difficult to do? 
> 
> the soundex module (or a similar technique, tuned to
> your dictionary) could be what you need:
> 
> import soundex
> print soundex.sound_similar("fredrik", "frederick")
> 1
> print soundex.sound_similar("fredrik", "hans")
> 0

Folks,

In my opinion, Soundex is generally far too hit-and-miss to be used in 
what is essentially a simple spelling-and-typo-correction application.

The basic problem is that it gives a yes-no answer. The proportion of 
false-positives and false-negatives will usually be found to be too 
high.

"Standard" soundex uses a four-byte key. I once had to repair the 
customer search component of an application --- among several 
sillinesses, it was using soundex(surname) as the sole method of 
grouping. This gave ludicrous results like:
-- search for "Mousaferiadis" and get one useful hit PLUS bucket-loads 
of "McPherson" (false-positives)
-- search for "Moudaferiadis" (single-key typo) and get no hits even 
though "Mousaferiadis" was in the database (false-negative)

Here are some illustrative results from the Python-supplied soundex 
module, which generates a 6-byte key:

>>> soundex.get_soundex("McPherson")
'M21625'
>>> soundex.get_soundex("Mousaferiadis")
'M21632'
>>> soundex.get_soundex("Moudaferiadis")
'M31632'
>>> soundex.get_soundex("Mousaferiadia")
'M21630'

This module is incorrect in its handling of multiple occurrences of the 
same code:
Bug #1 removes instances of the same code when they are separated by 
vowels (or W, H, Y).
Bug #2 has the inverse problem when the input starts with a doubled 
consonant.
>>> soundex.get_soundex("Mississippi")
'M21000' # Should be M221... (bug #1)
>>> soundex.get_soundex("Lloyd")
'L43000' # Should be L300... (bug #2)
>>> soundex.get_soundex("Loyd")
'L30000' # OK
>>> soundex.get_soundex("Lukasiewicz")
'L20000' # Should be L222... (bug #1)
>>> soundex.get_soundex("Lissajous")
'L20000' # Should be L222... (bug #1)

Reference: Knuth, D.E. "The Art of Computer Programming", vol 3, page 
392.

There have been various attempts to produce a better phonetic code e.g. 
NYSIIS code, Metaphon, ... but they still give a simplistic yes/no 
answer.

In Hans' application, a method which searched the whole vocabulary 
(small size in his case) and used a similarity score based on edit 
distance (a.k.a. Levenshtein distance) is what I would recommend --- 
see posting by Magnus Lie Hetland --- and maybe a follow-up or two :-)

HTH,
John




From paul at prescod.net  Fri Aug 20 18:07:53 1999
From: paul at prescod.net (Paul Prescod)
Date: Fri, 20 Aug 1999 22:07:53 GMT
Subject: Man pages would be cool.
References: <37BC6986.A66B044E@starmedia.net>
		<199908191634.KAA01333@jhereg.perl.com>
		<14268.21198.934117.241668@anthem.cnri.reston.va.us>
		<14268.23007.239507.166209@weyr.cnri.reston.va.us> <y0j7lmq44wd.fsf@vier.idi.ntnu.no>
Message-ID: <37BDD1B9.99588CB@prescod.net>

Magnus L. Hetland wrote:
> 
> "Fred L. Drake, Jr." <fdrake at acm.org> writes:
> 
> > Barry A. Warsaw writes:
> [...]
> >   Manpages have never been requested, unless you consider Tom's
> > complaint a request.  I've received nothing at
> > python-docs at python.org.  Certainly not a contributed conversion
> > script.
> 
> Hm. Could I request them now? I think it might be kinda nice to
> have... :)

I was very distressed to load up my Red Had system and find Python
effectively invisible. It should be available in man pages and IDLE
should be in the GUI menus. (obviously two very different problems)

 Paul Prescod



From dbparker at nortelnetworks.com  Wed Aug 18 08:44:26 1999
From: dbparker at nortelnetworks.com (Donald Parker)
Date: Wed, 18 Aug 1999 08:44:26 -0400
Subject: Tkinter support for toolbars
Message-ID: <7pe9it$4nq$1@bmerhc5e.ca.nortel.com>

I'm interested in reusing a python/tkinter toolbar (target system W9x/NT)
that has the look and feel of the toolbars supported by most windows
applications.
Anyone have an example of an implementation of this?





From 71351.356 at compuserve.com  Sun Aug  1 00:31:22 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Sun, 1 Aug 1999 00:31:22 -0400
Subject: 1.5.2 for: else:
References: <Pine.LNX.4.04.9907271632450.9849-100000@emerald.netskate.ru> <199907272208.AAA06264@axil.hvision.nl>
Message-ID: <7o0ipm$s18$1@ssauraab-i-1.production.compuserve.com>

ALL:
    I'm a newbie so I may be out of line here, but the for: else: construct
is pretty bad.  It just does not read well, and implies the incorrect
operation; it reminds me of  some of the weird FORTH stuff.   Why not use
another word for this and not overload 'else'?   Maybe for: also:, or
something more in line with the intended purpose.   Sure there may be good
reasons for else that I (newbie) do not know, but I thought Python was
supposed to enable ease of reading, not only writing (ala Perl).

TC Mits.


Hans Nowak <ivnowa at hvision.nl> wrote in message
news:199907272208.AAA06264 at axil.hvision.nl...
>
> On 27 Jul 99, Oleg Broytmann wrote:
>
> > > Loop statements may have an else clause; it is executed when the loop
> > > terminates through exhaustion of the list (with for) or when the
> > > condition becomes false (with while), but not when the loop is
> > > terminated by a break statement. [..]
> > [skip]
> > > Else clauses get executed on normal loop exit, even if the loop was
> > > 'empty'. It does not get executed when you 'break' out of a loop.
> >
> >    Aha, thanks. I expected "else" to be executed when "for" not executed
> >    on
> > empty list. Wrong assumption.
>
> This one has bitten me too in the past. One of the few non-intuitive
> parts of Python, methinks. :(
>
> --Hans Nowak (zephyrfalcon at hvision.nl)
> Homepage: http://fly.to/zephyrfalcon
> Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
> The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/





From mnot at mnot.net  Sat Aug 28 08:53:22 1999
From: mnot at mnot.net (Mark Nottingham)
Date: Sat, 28 Aug 1999 22:53:22 +1000
Subject: A Plea For A Return To Sanity!!!
References: <000301bef12c$1bfd6940$ad2d153f@tim>
Message-ID: <0fbd01bef154$4f7ab020$0301a8c0@mnot.net>

> python-folk-are-the-best-buddies-psu-dues-can-buy-ly y'rs  - tim

I don't know why, but I'm suddenly consumed with the desire to pay my PSA
dues... Hmmm...

Seriously, I've had a printed-out PSA application sitting under a pile of
paper on my desk for the better part of a year. Why? Well, have you ever
tried to post a check (or as they say here, cheque) from Australia to
Reston, VA? It's a pain, unreliable, and financially challenging.

Has anyone over there looked into one of the online payment services like
Kagi, etc? It would make things a lot easier, and you might get a few more
members. Like me. They take a few percent, but it's pretty reasonable.

Just a thought...

and-exactly-what-IS-the-connection-between-Python-and-the-Ebola-virus-ly
yours,

--
Mark Nottingham  Melbourne, Australia
http://www.mnot.net/




From jstevens at basho.fc.hp.com  Fri Aug 13 13:04:12 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: Fri, 13 Aug 1999 11:04:12 -0600
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <934326918snz@vision25.demon.co.uk>
Message-ID: <37B4500C.276AD22B@basho.fc.hp.com>

Phil Hunt wrote:
> >         6) what do you want?
> 
> C-style comments
> 
> sorted dictionary loops, eg:
> 
> for k, v in dict.sortedValues():
> 
> where this is sorted by keys.

If this last one is what you want, why not use either my B-Tree
container class, or my Trie container class?

Either one gets you reasonably fast mapping/unmapping, as well
as sorted keys.

John S.



From ionescu at pik-potsdam.de  Tue Aug 24 10:09:10 1999
From: ionescu at pik-potsdam.de (Cezar Ionescu)
Date: Tue, 24 Aug 1999 16:09:10 +0200
Subject: Newbie question about Re
References: <37C29893.D7B2C482@gnet.tn>
Message-ID: <37C2A786.262B78CD@pik-potsdam.de>

MB wrote:

> I just want to split a string of n x (p characters) into a list of
> p_length strings using the Re module
> Example: Assume p=5 the string '12345ABCDE12345IJKLM' must be split in
> ['12345','ABCDE','12345','IJKLM'']
> The way I  find is to make a replace before splitting:
>

I'm not sure I understand. why use re? Why not:

str =  '12345ABCDE12345IJKLM'
l = []
i = 0

while i < len(str):
	l.append(str[i:i+5])
	i = i+5

This gives the list you wanted.

All the best,
Cezar.



From greg at cosc.canterbury.ac.nz  Wed Aug 11 21:16:28 1999
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 12 Aug 1999 13:16:28 +1200
Subject: Macro preprocessor.
References: <slrn7r291c.137.kaz@ashi.FootPrints.net>
Message-ID: <37B2206C.CB010958@cosc.canterbury.ac.nz>

Sounds useful, although there are a lot of C-like things
in there (#include, braces, etc.) that would look somewhat
out of place in a Python program.

A truly Pythonic macro processor would use indentation for
delimiting macro definitions, and would use the existing
module system for managing namespaces, etc.

Also, what about line numbers? As far as I know, Python
has no equivalent of a #line directive, so keeping error
messages in sync with the source file is going to be
rather difficult.

-- 
Greg Ewing, Computer Science Dept, | The address below is not spam-
University of Canterbury,	   | protected, so as not to waste
Christchurch, New Zealand	   | the time of Guido van Rossum.
greg at cosc.canterbury.ac.nz



From nhead at houbits.com  Tue Aug 24 06:16:13 1999
From: nhead at houbits.com (Nigel Head)
Date: Tue, 24 Aug 1999 12:16:13 +0200
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan>
Message-ID: <37C270ED.89F59ABD@houbits.com>

My personal vote is to not propose to meddle with "pass":

1) as others have previously pointed out it is a great aid to
readability (one of the big points for python in my mind)

2) I often (well sometimes) generate code automatically and the "pass"
statement is something that I put in a lot when I have to deal with
unused options of a given code skeleton

3) I don't really mind using "pass #" as a comment-it-out string now and
again

4) I can't imagine that the time machine has ever changed this facet of
the syntax anyway

Nigel




From bwarsaw at cnri.reston.va.us  Sun Aug 22 18:16:51 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Sun, 22 Aug 1999 18:16:51 -0400 (EDT)
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
	<7ple6r$kf0@ds2.acs.ucalgary.ca>
	<37bf6144.1666109@news.freeserve.net>
Message-ID: <14272.30419.435335.609046@anthem.cnri.reston.va.us>

>>>>> "TD" == Toby Dickenson <htrd90 at zepler.org> writes:

    TD> Cool, I never knew that worked with other sequences too!

    TD> I guess "_tuple_ unpacking" isn't the best name for this
    TD> anymore ;-)

Right.  This changed circa 1.5 alpha 4, about the same time I added
class exceptions.  It actually works with any PySequence object
("sequence unpacking"), but lists and tuples get the fast
implementation.

-Barry



From gherman at my-deja.com  Wed Aug 11 07:39:16 1999
From: gherman at my-deja.com (Dinu C. Gherman)
Date: Wed, 11 Aug 1999 11:39:16 GMT
Subject: ANN: "Das Python Tutorium" - Public Review Release
Message-ID: <7ornd4$212$1@nnrp1.deja.com>

Hello,

for those who don't read comp.lang.python.announce: this is
a "short announcement", saying that the German translation
of the "official" Python tutorial by GvR is now available
for public review. Please check the following URL to down-
load it in one of various formats:

   http://starship.python.net/crew/gherman/propaganda/tut-de/

Better even, read the full announcement you will find there
or in comp.lang.python.announce before doing so to get some
more details.

Dinu

................................................................
"An average of more than 15 % of adults in 12 industrialized
countries are functionally illiterate; in Ireland, the United
Kingdom and the United States, the rates are over 20 %."

  (The State of the World's Children 1999,
   UNICEF, http://www.unicef.org/sowc99)




Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From WESTER at ilt.fhg.de  Mon Aug  2 09:48:42 1999
From: WESTER at ilt.fhg.de (Rolf Wester)
Date: Mon, 02 Aug 1999 15:48:42 +0200
Subject: TclError: Can't find a usable init.tcl
Message-ID: <37A5A1BA.77BDF2B3@ilt.fhg.de>

Hi,

I installed Tcl8.1, Tk8.1 and Python1.5. Trying

tk=Tkinter.Tk()

I get the error message
...
TclError: Can't find a usable init.tcl in the following directories:


This probably means that Tcl wasn't installed properly.

The directory list which is searched for init.tcl is obviously empty. My
init.tcl is in the directory /usr/local/lib/tcl8.1. How can I make
Python to search in this directory?

I would be very appreciative for help.

Rolf

------------------------------------------------------------
# Rolf Wester
# Fraunhofer Institut f. Lasertechnik
# Steinbachstr. 15, D-52074 Aachen, Germany.
# Tel: + 49 (0) 241 8906 0, Fax: +49 (0) 241 8906 121
# EMail: wester at ilt.fhg.de, WWW: http://www.ilt.fhg.de





From jon at rdt.monash.edu.au  Thu Aug 26 22:00:05 1999
From: jon at rdt.monash.edu.au (Jonathan Giddy)
Date: 27 Aug 1999 02:00:05 GMT
Subject: fully qualified hostname ?
References: <199908261822.AA01099@ch1d2833nwk> 	<595C6D6DD3292BB3.DCF1BAEEAE3E421A.EFDF99C4B418A66D@lp.airnews.net> <199908262103.AA01157@ch1d2833nwk>
Message-ID: <7q4rf5$uio$1@towncrier.cc.monash.edu.au>

Ted Horst <Ted.Horst at wdr.com> writes:
>> >
>> >What is the best way to get the fully qualified hostname in python (ie
>> >hostname.domainname) ?
>> >
>> >A portable solution would be ideal, but I would settle for Unix only.

Under Unix, I get the best (i.e. least surprising) results using the 
following gethostname function:


import socket

def getdnsnames(name):
    d = socket.gethostbyaddr(name)
    names = [ d[0] ] + d[1] + d[2]
    return names

def resolve(name):
    names = getdnsnames(name)
    for dnsname in names:
        if '.' in dnsname:
            fullname = dnsname
            break
    else:
        fullname = name
    return fullname

def gethostname():
    fullname = socket.gethostname()
    if '.' not in fullname:
        fullname = resolve(fullname)

    return fullname

--
jon at dstc.edu.au



From aconstott at aol.com  Tue Aug 17 23:45:21 1999
From: aconstott at aol.com (aconstott at aol.com)
Date: Tue, 17 Aug 1999 21:45:21 -0600
Subject: $1/MINUTE PHONESEX W/ REAL PEOPLE LIKE YOURSELF 57417
Message-ID: <170899214521@aol.com>

Wanna meet someone tonight?

        CALL NOW:    
1-800-750-GIRL (4475)

* Guys: $1 per minute
* Girls: Totally FREE

These girls are not phonesex workers. They are horny girls who are on the line cuz it's free phonesex for them. You are free to arrange to meet anyone you meet on this line. Give it a try!





Sc2)\_>I



From tim_one at email.msn.com  Sat Aug 21 04:09:21 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 21 Aug 1999 04:09:21 -0400
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <37BCB288.B4476B8B@vision.arc.nasa.gov>
Message-ID: <000301beebac$792199a0$0c2d2399@tim>

[Christopher Browne]
> Is it so much work to come up with a half-reasonable card shuffling
> algorithm?
>
> The one I've always seen cited as providing decent behaviour looks
> thus:
>
> for (i = 0; i > n; i++) {  /* surely intended "i < n" */
>    swap (item[i], item[random(n)]);
> }

[Chad Netzer]
> A list of n items has n! permutations.  The algorithm you propose
> will generate  n**n possible execution paths (for lack of a better
> term), which is not evenly divisible by n! .  So your variation
> does not produce all possible permutations with equal probability.
> ...
> PS.  if I f*cked up, be gentle. :-)

No, that was an excellent argument, Chad!  Cut straight to the heart of it
with no need to think <wink>.  For n=1 and n=2, Christopher's algorithm
works fine, and for all n the *last* position in the generated permutations
has the right distribution.  For all other cases, the output element at
position i is most likely to be the element that was at input position i+1.

I think this algorithm persists because it's obvious that, on the i'th step,
every element has an equal chance of ending up at position i.  So, by
induction <wink>, all permutations are equally likely in the end.  That
argument doesn't make sense, but the flaw is hard to see.

Hint for the curious:  after the first step, every element is indeed equally
likely to be in the first position.  After the second step, the same is true
of the second position.  But after the second step it's no longer true that
every element is *still* equally likely to be in the first position; in
fact, there are n ways for the original first element to be unchanged (one
of which swaps it twice), 2*n-2 ways for the original second element to be
in the first position after the second step, and n-1 ways for each of the
remaining n-2 elements.

asymmetry-hides-in-symmetry-like-tears-in-the-rain-ly y'rs  - tim





From robin at jessikat.demon.co.uk  Mon Aug 23 07:28:35 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 23 Aug 1999 12:28:35 +0100
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk>
 <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu>
 <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu>
 <37c126a6$0$813@newsreader.alink.net>
Message-ID: <wuvtICAjBTw3Ewxa@jessikat.demon.co.uk>

In article <37c126a6$0$813 at newsreader.alink.net>, xah <xah at weborder.com>
writes
>
>Tom Christiansen wrote in message <37b414d7 at cs.colorado.edu>...
>>     [courtesy cc of this posting mailed to cited author]
>>
>>Legibility is purely subjective.
>>
>>How legible do you find Greek, Arabic, Russian, or Chinese?  And what
>>does the character set actually have to do with any inherent complexity
>>of the grammar or richness of expressive vocabulary?
>>
>>Answer: nothing whatsoever.
>
>
>I'd like to extend what Tom Chrisiansen has said.
>
...
>On a personal note...  Do you know what a computer science degree is these
>days? They require you understand shits like lambda calculus, logic, graph
>theory, combinatorics, or even algorithms?? All for what? All the Perl
>programers I know doesn't understand any of these, and they all are making
>big bucks in fortune 500. I'm not ashamed that I dropped out of school. I
>hope that Larry and Tom's teachings will eventually make such subjects
>disappear for good.
>
presumably they'll stop schools in Kansas teaching bible studies in
favour Larry & Tom readings.
>My humble $0.02.
>
> Xah
> xah at best.com
> http://www.best.com/~xah/PageTwo_dir/more.html
> "Your nonsense is no better than mine."   --Mathematician of metaland.
>
>
>
>

-- 
Robin Becker



From p.t at iname.com  Sun Aug  8 05:01:28 1999
From: p.t at iname.com (Peter Torstenson)
Date: Sun, 8 Aug 1999 06:01:28 -0300
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk>
Message-ID: <7ojgv8$2drq@enews1.newsguy.com>

Phil Hunt <philh at vision25.demon.co.uk> wrote in message
news:934075710snz at vision25.demon.co.uk...
> I have two python modules, both of which need to refer to classes
> defined in the other one. How do I do this?
>
> I am currently putting at the start of each module a line of the form:
>
>    from theOtherModule import *
>
> However, this is not working. My modules -- greatly simplified -- look
> like this:
>
>    #bak.py
>    from hhh import *
>    ti = AnHhhClass()
>    print ti
>
> and:
>
>    #hhh.py
>    from bak import *
>    class AnHhhClass: pass
>
> When I run ``python bak.py'', I get the error message
> ``NameError: AnHhhClass''.
>
> --
> Phil Hunt....philh at vision25.demon.co.uk

Referring to something in an other module you have to put the modulename in
front:
Like in bak.py: "ti=hhh.AnHnnClass()"

/Peter





From jstevens at basho.fc.hp.com  Fri Aug 13 22:53:49 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 19:53:49 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132250.QAA06265@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 4:50 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cc2d@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :OTOH, Python is perfect for the OO trained programmer who is part of
> :a team, or where the code will have to be maintained over a long
> :period of time by many different people.
> 
> As is Perl.

That turns out not to be the case.  In three projects that started
out in Perl, two were eventually converted to Python, for just that
reason: Python is the much better choice for those kinds of tasks.

> A sound Perl OO module designed by a competent programmer
> is certainly up to fitting that bill.

That turns out, again, not to be the case.  Unless, of course, you
define competent in a self-referential fashion.

John S.




From robin at jessikat.demon.co.uk  Mon Aug  9 08:59:39 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 9 Aug 1999 13:59:39 +0100
Subject: Licenses...
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com>
 <37AA804D.9A1317C0@houbits.com> <934115482snz@vision25.demon.co.uk>
 <37ae686e.59511102@news.ricochet.net> <37AE9AC3.34007C4E@lemburg.com>
 <37AE9B42.4DF40758@lemburg.com>
Message-ID: <SpY1dCA7Ctr3EwsH@jessikat.demon.co.uk>

In article <37AE9B42.4DF40758 at lemburg.com>, M.-A. Lemburg
<mal at lemburg.com> writes
...
>If you link against a GPLed product (even if it's done dynamically!),
>the whole construction (your product + the GPLed one) automatically
>fall under the GPL. This means that you can still sell and distribute
>your product, but only if you include the full source code of
>the GPLed product
>*plus* your product.
>
Oh dear; I wrote this extension called dll for Tcl that allows tcl code
to link to arbitrary dll's. The source code for the extension is
available. Must I distribute source for all known GPL'd products which
can be linked to tcl using this extension. As it happens this thing is
mostly used by win32ers, but it has run under Linux.
>The LGPL is a little different: it allows you to link against the
>LGPLed product without the need to ship your source code. You still
>have to include the source code of the LGPLed product you're linking
>against though (including any changes that you have made to it).
>
>If you are using Python extensions written by different authors,
>be sure to check their license statement. In case you have any
>doubts get in touch with them before using their code. Most 
>extensions are under very similar licenses as Python itself,
>but some a GPLed and other LGPLed. Also, some are not valid for
>commercial use meaning that you'll have to negotiate a special
>license with the author to be able to use in your code. This
>usually involves paying a license fee.
>
>Hope this clarifies things a little. More infos on GPL and LGPL
>should be available on www.gnu.org. If you are still unsure what
>is legal and was is not, go on the safe side and contact your
>lawyer to get things straight.
>

-- 
Robin Becker



From cjc26 at nospam.cornell.edu  Thu Aug 19 19:49:41 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 19 Aug 1999 23:49:41 GMT
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <y0jiu6edoa9.fsf@vier.idi.ntnu.no> <xj0d7wl9acd.fsf@cyclone.harvard.edu> <slrn7rm1ol.s1.cjc26@tankgrrl.bridget.mindriot.net>
Message-ID: <slrn7roo2j.mt.cjc26@tankgrrl.bridget.mindriot.net>

Pada 18 Aug 1999 23:16:42 GMT, Cliff, or a close facsimile bilang:
| | Each of those actions is O(N) so the whole loop is O(N**2).
| 
| Nope, it should still be O(n).

Oops..I was wrong.  It >is< O(n^2), not O(n).
My only excuse is that it was Wednesday, and Wednesdays don't like me :)


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From bowman at montana.com  Mon Aug  9 23:39:39 1999
From: bowman at montana.com (bowman)
Date: Mon, 09 Aug 1999 21:39:39 -0600
Subject: looking for py/database ideas
References: <m2so5slkdj.fsf@netmeg.net> <kNKr3.879$gO1.33509@news2.giganews.com> <m2n1w0l4al.fsf@netmeg.net>
Message-ID: <37AF9EFB.44E10EA9@montana.com>

Les Schaffer wrote:
> 
> Chris Browne wrote:
> 
> not sure my friend is so much interested in linux vs. NT/9X. but he
> may be interested in an alternative approach to the database solution,
> and if that takes him away from Windows, "so be it".

There is a package for mySQL, though mySQL might not be professional
(pricey) enough to impress your friend. It would be quite easy to
develop an interface to db2, which is certainly professional and robust
enough, and is available for AIX, Linux, NT, and other platforms. Be
advised db2 tends to be huge and resource hungry, and may require a full
time administrator for non-trivila applications.

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From fredp at multimania.com.nospam  Tue Aug 31 08:58:36 1999
From: fredp at multimania.com.nospam (Fred Pacquier)
Date: 31 Aug 1999 12:58:36 GMT
Subject: Newbie (read: pot. stupid) question on objects/namespaces
Message-ID: <8E339DC2FPaCmAnRDLM@news.dial.oleane.com>

Hi,

I tried looking this up in the FAQ/tutorial and failed, but maybe it's a 
terminology problem.

I'd like to know if it is possible, given the name of a "data attribute", 
to find out the name of its "owner", ie the instance (namespace) where it 
has been _created_ (the instance, not the class). This is probably unclear, 
so to illustrate :

def func(a_thing,...)
    	...

class foo:
    	self.thing = ...

class bar:
    	self.a_foo = foo()
    	func(self.a_foo.thing,...)

Here, 'func' (might be anywhere, top-level, another class, in the same 
module if it helps...) receives the data attribute 'thing' of instance 'a-
foo' of class 'foo'. This instance was created (by ? inside ?) class bar, 
or (probably) an instance 'a_bar' of class 'bar'.

The question is: given the simple call above with only the name as a 
parameter, is there a (legal) way for function 'func' to determine that 
'a_thing' actually (lives in ? belongs to ?) 'bar' (or 'a_bar') ?

Sorry if the question is really sick :-)

-- 
YAFAP : http://www.multimania.com/fredp/



From jstevens at basho.fc.hp.com  Fri Aug 13 23:01:12 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:01:12 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132255.QAA06466@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 4:55 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cde8@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :>     @array = ("fred", "wilma", "barney", "betty");
> :>     %hash  = @array;
> :> Is the same as:
> :>     %hash  = ("fred", "wilma", "barney", "betty");
> :> Because the list stored in the array will have its key-value pairs
> :> re-interpreted when this is stored into the hash.
> :
> :Resulting, again, in someting confusing to the programmer who
> :hasn't any idea of what the above does, and that is still
> :somewhat confusing even to an intermediate Perl programmer.
> :
> :I can *guess*, from my Perl experience, that this creates a hash whose
> :keys are the elements of the list, and that the data elements
> :mapped to those strings are . . . wait a minute, I'm not absolutely
> :sure.
> :
> :Let me try it. . . Ok, the data elements are set to either an integer
> :zero, or a string whose contents are the zero character, or a floating
> :point zero.
> 
> Goodness no, it doesn't mean that at all.  Key-value pairs, dude.

??? 

When I tried it, then unmapped back, I got some kind of zero.

So much for DWIM.  Perl didn't do what I meant, there, and I couldn't
even guess, I had to try it.

And had I run across this idiom in a Perl program, I would have had harsh
words for the author, unless the code had been heavily documented.

John S.




From kaz at ashi.FootPrints.net  Wed Aug 11 03:20:12 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Wed, 11 Aug 1999 07:20:12 GMT
Subject: Macro preprocessor.
Message-ID: <slrn7r291c.137.kaz@ashi.FootPrints.net>

Hi folks. I wrote a macro preprocessor that can be used with a language that
uses indendation for statement or declaration grouping, because it is
intelligent about dealing with leading whitespace. 

http://users.footprints.net/~kaz/mpp.html

I haven't done work on it in a long while and I need some motivation to to add
features to it to make it more useful. (It needs conditionals, numeric
variables and operations, iteration constructs). It would help, like, to
have some users. :)

I know jack about Python, but I was partially thinking of you guys when I added
whitespace intelligence to the list of requirements.  (That and the desire to
have a preprocessor tool that doesn't leave obvious artifacts usually caused by
preprocessing, such as broken formatting, and gives you great control over
whitespace).

In this macro preprocessor you can do something like this:

    #def snippet1 {
	line1
	line2
	line3
    }

    #def snippet2 {
	line4
	    #snippet1
	line5
    }

    def func(x)
	#snippet2

and the output will be:

    def func(x)
	line4
	    line1
	    line2
	    line3
	line5

as you would intuitively expect, rather than coming out in some horrible mess
as would happen with a traditional C preprocessor utility, or something like
m4, either of which would make dog's breakfast out of, say, a Python program. 

There are other features im MPP. One noteworthy feature is that it supports C++
style nested namespaces:

    #namespace html {
	#def url(link, text) {<a href="http://#link">#text</a>}
    }

    #using html.url
    #url(www.slashdot.org, slashdot)

    #comment { or full scope resolution }
    #html.url(www.freshmeat.net, freshmeat)

    #comment { create alternate name for something }
    #alias web_stuff html
    #web_stuff.url(www.kernel.org, kernel)

    #comment { and other cheap symbol-table tricks. :) }

With namespaces you can create macro libraries that don't clash.  Speaking of
which, there is a fully blown #include directive, with "..." and <...> header
searching logic that can be influenced with command-line options.  There is a
#once directive to suppress multiple inclusions of same header.  Great
diagnostics that pinpoint the line number and file where an error was caused,
accompanied by a dump of the whole #include chain that led up to the error, if
it was in a header. Oh, and did I mention that you can dynamically modify the
lexical character mapping, similar to the way codes can be redefined in TeX? If
you don't want the # character to introduce directives, you can change it to
something else on the fly. (If you change it within a header, the change won't
automatically be propagated to the host file that included it.)

    #code leftbrace [
    #code rightbrace ]
    #code escape %

    %def foo [ bar ]	

Okay, enough. :) I'll probably get hate mail now about being off topic.



From moshez at server.python.net  Fri Aug 20 07:08:16 1999
From: moshez at server.python.net (Moshe Zadka)
Date: Fri, 20 Aug 1999 07:08:16 -0400 (EDT)
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
In-Reply-To: <37BC7014.33B6FFF6@compaq.com>
Message-ID: <Pine.LNX.4.10.9908200706500.15847-100000@starship.python.net>

On Fri, 20 Aug 1999, Greg Ewing wrote:

> Martijn Faassen wrote:
> > 
> > Funny, I read that 'the way g-d indented'. :)
> 
> Hey, what a great signature that would make!
> 
> Python: Programming the way g-d indented.

Make that:
Python: Programming the way g--d- indented.




From rceci at adams.com  Mon Aug  2 11:04:35 1999
From: rceci at adams.com (rceci at adams.com)
Date: Mon, 02 Aug 1999 15:04:35 GMT
Subject: HELP: ImportError on HPUX
Message-ID: <7o4c1s$n7d$1@nnrp1.deja.com>

Hello,
I have a c++ library that runs great on HPUX (10.20) and IRIX (6.4), in
a native (ie non-Python) environment.  I have also been successful at
getting the library to import w/o problems on IRIX, as a SWIGged module
in Python.  Now, I am having difficulty getting the library to import on
HP.  I found an old reference to C++ modules in HP needing
"Python/importdl.c" to have shl_load changed to cxxshl_load and what
not.  I am not sure that is still valid, or if I have some other build
problem, specific to HPUX.  I am using Python 1.5.1. and SWIG1.1p5. When
I import the python library I
get the following errors:

Python 1.5.1 (#1, Jul 22 1999, 13:51:48) [C] on hp-uxB
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Traceback (innermost last):
File "/vobs/release//run_python.py", line 7, in ?
  import MDIreg
File "/vobs/release//python/lib/hpux10/python1.5/MDIreg.py", line 2, in
?
  import MDIregc
ImportError: Failed to load
/vobs/release//python/lib/hpux10/python1.5/MDIregcmodule.sl
>>>

======

adding -v to the python invocation tells me more:

# /vobs/release//python/lib/hpux10/python1.5/exceptions.pyc has bad
mtime
import exceptions # from
/vobs/release//python/lib/hpux10/python1.5/exceptions.py
# can't create /vobs/release//python/lib/hpux10/python1.5/exceptions.pyc
Python 1.5.1 (#1, Jul 22 1999, 13:51:48) [C] on hp-uxB
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
# /vobs/release//python/lib/hpux10/python1.5/MDIreg.pyc matches
/vobs/release//python/lib/hpux10/python1.5/MDIreg.py
import MDIreg # precompiled from
/vobs/release//python/lib/hpux10/python1.5/MDIreg.pyc
shl_load /vobs/release//python/lib/hpux10/python1.5/MDIregcmodule.sl
/usr/lib/dld.sl: Unresolved symbol: __StaticCtorTable_Start (data)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: __ZStaticCtorTable_End (data)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: __StaticCtorTable_End (data)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: __ZStaticCtorTable_Start (data) from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: U_update_state_vector (code)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: U_get_shLib_unw_tbl (code)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: U_resume_execution (code)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: U_get_previous_frame_x (code)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: U_get_unwind_table (code)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: U_get_shLib_text_addr (code)  from
/usr/lib/libCsup.1
/usr/lib/dld.sl: Unresolved symbol: U_get_unwind_entry (code)  from
/usr/lib/libCsup.1
shl_findsym initMDIregc
import MDIregc # dynamically loaded from
/vobs/release//python/lib/hpux10/python1.5/MDIregcmodule.sl
Contacting Registry ...
Bus error

By the way, I am using static C++ constructors in a few cases and am not
sure that is part of the problem.  I am using:

Python 1.5.1, SWIG 1.1p5, on HP-UX 10.20, compiler levels:
HP aC++ B3910B A.01.18
HP aC++ B3910B A.01.12 Language Support Library

Many Thanks


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From hnowak at cuci.nl  Sun Aug 29 14:06:59 1999
From: hnowak at cuci.nl (Hans Nowak)
Date: Sun, 29 Aug 1999 18:06:59 GMT
Subject: Python for Delphi current?
In-Reply-To: <7qbn1j$4jp$1@nnrp1.deja.com>
References: <7qbn1j$4jp$1@nnrp1.deja.com>
Message-ID: <199908291808.UAA29785@dionysus.fw.cuci.nl>

On 29 Aug 99, at 16:27, sfranke at my-deja.com wrote:

> Does the Python for Delphi Package work with
> Python 1.5.2 and Delphi 4(5)?

I tried to build it for Delphi 4, but failed. Then again, I do not 
know that much about packages and components and such in Delphi. :-/ 
So I don't know how valuable my information is.


--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/



From news at dorb.com  Thu Aug 26 11:30:53 1999
From: news at dorb.com (Darrell)
Date: Thu, 26 Aug 1999 11:30:53 -0400
Subject: Is there a pattern...
References: <y0jg116vaoa.fsf@vier.idi.ntnu.no>
Message-ID: <G4dx3.1625$v6.105061@typhoon.nyroc.rr.com>

Python doesn't support private anything. Although mxProxy may help in this
case. http://www.lemburg.com/python/

Another goofy approach might be:
Untested:

import new

def __unreg(self):
    """
    Do some unregistering here
    """

class Reg:
    def register1(self)
        """ could have returned a lambda also """
        return new.instancemethod(__unreg,self,Reg)

--Darrell






From claird at starbase.neosoft.com  Fri Aug 27 10:01:51 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 27 Aug 1999 14:01:51 GMT
Subject: Why do we call python a scripting language?
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com>
Message-ID: <D934E87AB6D3A547.B15B4CC9AD77B864.400826439A461ACE@lp.airnews.net>

In article <8E2F958DCPaCmAnRDLM at news.dial.oleane.com>,
Fred Pacquier <fredp at multimania.com.nospam> wrote:
			.
			.
			.
>You have something there. Maybe another way would be to showcase the 'real' 
>applications and their strengths, preferably in ways people can relate and 
>compare to similar 'traditional' apps. Unfortunately it seems (to me) that 
The "Regular Expressions" column <URL:http://
www.sunworld.com/sunworldonline/common/swol-backissues-columns.html#regex>
(incidental remark:  I'll probably do up a more
interesting index sometime in September) makes
a point of featuring an example of "scripting
in the real world" in the first installment
each month.  Suggestions are welcome, of course.
>such desktop, user-oriented apps in Python are still scarce ; much of the 
>industrial-strength (like BSCW or Zope) happens on the backend and is Web-
>oriented. Impressive for the initiate, but abstract and hard to marvel over 
>for the profane. The one app that 'did it' for me, curiously, is PySol. 
"Regular Expressions" has been emphasizing
everything *except* the desktop, although Sep-
tember will be an exception.  It's interesting
to learn what (like PySol) makes an impression
on people.
			.
			.
			.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From m.faassen at vet.uu.nl  Mon Aug 23 04:52:56 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 23 Aug 1999 08:52:56 GMT
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer)
References: <001601beeaef$e795f380$712d2399@tim> <37bd4385@cs.colorado.edu>
Message-ID: <7pr258$mdg$1@newshost.accu.uu.nl>

Tom Christiansen <tchrist at mox.perl.com> wrote:
> I remember my frutration as a flaming C advocate (you didn't really
> think I started with Perl, did you? :-) getting a job at a company
> that made US$75k Fortran compilers being told that Fortran would always
> be faster than C because the fc compiler could know more things about
> your code than the cc compiler could, and thus make better compile-time
> optimization to speed run-time.  I wonder whether Python simply hasn't
> matured as far as Perl has along those lines, or whether in fact there
> are inherent barriers, as there are between Fortran and C.  What do you 
> think?  I also notice that Python is bereft of compiler pragmata. 

In part it's inherent problems. Python is way dynamic and therefore hard
to speed up.

But, Pythoneers also know it can be done, by looking at other languages
that are dynamic, such as Smalltalk and Lisp, which have managed to gain some
quite massive improvements in speed.

Pythoneers are actively looking at ways to speed up Python code, so in part
it's also simply maturing. I heard Python *has* speeded up a lot since
earlier versions, in fact.

Regards,

Martijn




From mwh21 at cam.ac.uk  Fri Aug 13 02:44:02 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Aug 1999 07:44:02 +0100
Subject: solaris 7--popen2 crash
References: <XFMail.990812173656.nathan@islanddata.com>
Message-ID: <m3so5o2n9p.fsf@atrus.jesus.cam.ac.uk>

Nathan Clegg <nathan at islanddata.com> writes:

> This is the first time I've had a test crash on me.  I've installed python
> on linux, solaris, and HP-UX, but this is my first time on solaris 7. 
> During make test, I got this:
> 
> test test_popen2 crashed -- exceptions.AssertionError : 
> 
> Any ideas?  Missing libraries?  Incompatible platform?  Any more helpful
> information I can provide?  Worse case scenario: can I continue on using
> python as long as I dont' use popen2? :)

>From the output of "cvs log Lib/popen2.py":

revision 1.8
date: 1999/04/20 12:27:31;  author: guido;  state: Exp;  lines: +2 -1
Calling _cleanup() does not guarantee that all processes have
terminated; this makes the final assert in the self-test code fail if
the parent runs faster than the children.  Fix this by calling wait()
on the remaining children instead.

I.e. it's probably on non-problem.

HTH
Michael



From I.Clarke at strs.co.uk  Thu Aug 12 04:09:06 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Thu, 12 Aug 1999 09:09:06 +0100
Subject: Why is CGI module so complicated?
References: <cWYr3.1329$EG4.226668@news.shore.net> <NDBBIAPHEKGOBIHCBLIIMEDNCAAA.mcfletch@vrtelecom.com>
Message-ID: <37B28122.FF630C52@strs.co.uk>

> What about simply adding a __str__ method to these objects?
> That way for printing and '%s'%(object) substitution you get the "simple"
> response of returning the (likely de-escaped) string value stored in the
> field. Similarly, add a "list" method to get the values as a list (possibly
> of length one) (is there a special method name that the built-in list
> function calls? I could not find it.)

I think this is an excellent idea... can anyone see any problems with
it?

I think what we are trying to do here is retain the flexibility of the
current method, but prevent the more sophisticated functionality from
making it more difficult to do simple stuff (like, access a text field).

Ian.



From aahz at netcom.com  Mon Aug 23 11:12:50 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 23 Aug 1999 15:12:50 GMT
Subject: Authentication with httplib
References: <7pr09c$dag$1@news.netvision.net.il> <7prana$a0p$1@minus.oleane.net>
Message-ID: <7prodi$48n@dfw-ixnews3.ix.netcom.com>

In article <7prana$a0p$1 at minus.oleane.net>,
Mathieu Lecarme <zehp at francemel.com> wrote:
>
>you should rather use urllib :

No, you shouldn't, unless you're running an interactive process.  urllib
expects stdin to provide name and password.  We got bitten by this one;
because we don't care about authorization (yet) but do care about
redirection, we had to subclass fancy_url_opener() (or whatever the
exact spelling is).

If someone is running a batch process, zie will either need to do what
the original poster was doing (use httplib directly) or subclass
urlopen() to make it work in a batch environment.

I've actually got some other problems with urlopen() in that it doesn't
return the HTTP return code.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From wtanksle at dolphin.openprojects.net  Fri Aug 20 00:03:30 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 20 Aug 1999 04:03:30 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
Message-ID: <slrn7rpksg.e6c.wtanksle@dolphin.openprojects.net>

On 19 Aug 1999 08:21:55 -0700, Tom Christiansen wrote:
>Per your collective requests, here is one of the documents
>with my collective observations from learning Python.

A very cool document.  Looks like my observations about Perl, except I
didn't write anything down (too stupid-lazy).

>This article was posted to the Perl newsgroup, but I have duplicated
>the posting here without crossposting to avoid flame wars.

Good idea.  Pity it didn't work, eh?  ;-)

>--tom

>x / y                           Same for floats, but Perl doesn't do
>                                integer division here usually.  Python
>                                things 3/4 is 0 (and false), but 3.0/4.0 is 0.75

Q: can Perl be made to do integer division here if you declare both
variables as integer?  I seem to recall something like that (but it's been
a while).

>                                strings aren't numbers, nor are they
>                                autoconverted.

This is important to Python.  And Perl, of course.

>float(x)                        not applicable

Except for variables declared integer, right?

>pow(x,y)                        x ** y

x**y is also valid Python syntax.

>x << n, x >> n                  same, although Perl works differently if
>                                you go over 31 bits:  Python keeps giving 
>                                you 0, whereas Perl wraps.  1<<34 is 0 in 
>                                python but 4 in perl.

Facinating.  Is the Perl << actually a rotate, or is that a special case
for n > 31?

>~x                              same-ish, but see |&^ above for details.
>                                In particular, you can't negate bitstrings
>                                in python.

I'm not sure what bitstrings are, but Python's longs can certainly be
negated, and I'm not sure what more you can ask for without asking for
Common Lisp.

(Here's an idea: take Common Lisp, rename its functions, write a new
reader macro, and call it "Python 2.0"!)

>Python Construct                Perl Equivalent
>===========================================================================

>                                Note that Python doesn't support hash
>                                slices, which in Perl are simple:
>                                    @hash{k1,k2,k3} = (v1, v2, v3)

That's a good point, although I'm not sure what Python can do about it --
our slice semantics talk about gaps, and hashes don't have gaps.  I
suspect we'd have to call it something other than slicing (this seems
pretty reasonable).

>hash(x)                         huh?  "returns the hash value of the object,
>                                if any"

The key which a hash table would use to store and look up the indicated
data.

>hex(n)                          sprintf("%#x", n)
>                                Perl's hex() function is the opposite.
>                                It converts 22 into 37.  Python's hex
>                                function converts 37 into 0x22.  Oh my!

Now this is a cool difference, because it highlights a difference between
the languages.  hex in Python converts from an integral number to a
string; in Perl it converts from a number which is treated as a string to
a number (which can be treated as a string).

Amusingly, applying hex repeatedly in Perl results in an ever-increasing
number.

I think Perl is more than a little oddball here.  I tried to model what
was hapenning, but I couldn't figure it out.  Ah, here we go, Perl does
the equivalent of converting the number into a string using radix 10, and
then reading that string into an integer using radix 16.

This is exactly what you'd want if you'd just read a hex number in which
didn't begin with "0x", so the read system misinterpreted it as a decimal.

Am I right?  I'm pretty puzzled here...

>reload(module)                  First delete from %INC, then require the
>                                module again, as in:
>                                    delete $INC{"Module.pm"}
>                                    require Module;
>                                Note that Python doesn't directly support
>                                nested modules as in
>                                    require Dir1::Dir2::Dir3::Module;
>                                except via a disgustingly named hackaround
>                                module.

True for 1.4, not true after that.

And "ni!" isn't disgusting -- what's disgusting is what the times have
come to, that ruffians can walk about saying "ni!" to innocent old ladies.

>str(x)                          I thought this was the same as repr().

Similar, identical in many cases.  str() is more verbose (prettier) in
some cases.

>tuple(seq)                      n/a.  This is a nasty hack to deal with the
>                                fact that (1,2,3) is different from [1,2,3].
>                                Hm... maybe it's like this:
>                                    $aref = [1,2,3]; 
>                                    @list = @$aref;

By "nasty hack" you mean "obvious consequence", of course.  Just like
$obj->{$name} is an obvious consequence of interpolation.

Sorry, I'm getting editorial -- but then you tend to be very much so.

>file.flush()                    Either FH->autoflush(1) if you're using the
>                                aliasing modules, or else the klunky but
>                                fast:
>                                    $old_fh = select(FH);
>                                    $| = 1;
>                                    select($old_fh);
>                                Note that this just sets autoflushing
>                                on each subsequent output on that handle.

Q: the Python version merely flushes the file, while the Perl version
looks like it's actually setting a permanent flag in the file to flush
frequently.  Is that correct?

Python, OTOH, doesn't seem to have a way to set a file's flushing
properties -- you have to explicitly flush whenever you need it.  True?

>sys.exit(status)                exit(status) and no import is bloody
>                                necessary just to leave the program.
>                                But that Python does this by raising
>                                a SystemExit exception.  Perl can use
>                                END{} handlers to catch these and do
>                                at-exit processing, and Python doesn't.
>                                You have to roll your own.

Python uses "finally:" and "except SystemExit:" blocks to do that.  Is
that what you mean?

I seem to recall END blocks in Perl, although I never used them...  Don't
they remain active forever, like C's atexit function?  How do you
deactivate them?

>_exit(n)                                POSIX::_exit(n)

In Python _exit skips the atexit stuff.  True in Perl?

-- 
-William "Billy" Tanksley



From robin at jessikat.demon.co.uk  Mon Aug 23 09:47:29 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 23 Aug 1999 14:47:29 +0100
Subject: filtering a list
Message-ID: <yILNBDAxDVw3EwSE@jessikat.demon.co.uk>

I'm uncertain as to whether this kind of construct is well formed

R is a list of instances

for i in R:
  if i.id%2: del R[R.index(i)]

or is it better (assuming R holds the only references to the instances)
to do

R=filter((lambda x: x.id%2),R)
-- 
Robin Becker



From philh at vision25.demon.co.uk  Wed Aug 25 15:05:33 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Wed, 25 Aug 99 19:05:33 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <14274.51855.281578.240389@dolphin.mojam.com> <slrn7s6mfo.68l.cbbrowne@godel.brownes.org> <slrn7s6o94.im.irclark@latveria.castledoom.org>
Message-ID: <935607933snz@vision25.demon.co.uk>

In article <slrn7s6o94.im.irclark at latveria.castledoom.org>
           irclark at latveria.castledoom.org "Isaac" writes:
> On Wed, 25 Aug 1999 03:02:30 GMT, Christopher B. Browne 
> <cbbrowne at news.brownes.org> wrote:
> >This of course has the classic problem that
> >
> >  You can only usefully include things in "Parrot" that are
> >  expressible in all of the GUI systems, thereby reducing Parrot to a
> >  Lowest Common Denominator.
> >
> >That's not 100% true; there *could* be Parrot features that would be
> >ignored on platforms that did not support them.  Some danger lies down
> >that road...
> 
> As long as they weren't silently ignored,

Errors, and features not implemented on the particular backend
chosen, will be written to a log file, and to stdout.

> it wouldn't really be dangerous.
> The least common denominator for say GTK and motif is actually fairly
> useful. 

Indeed; most GUIs are based on CUA -- a 1987 specification -- and
there's been little innovation in the last 10 years. Tabbed dialogs
are an obvious exception to that rule.

> What really makes this kind of tool hard to write is the differing 
> component placement algorithms from one toolkit to the next.   You
> probably won't easily be able to support all of the options for a
> every toolkit.  Phil's syntax for parrot suggests a certain type
> of geometry manager.  Toolkits that won't support that manager
> will probably be the most difficult to support with Parrot.

This is true. If a GUI toolkit doesn't have an analog of rowLayout
or colLayout, a Parrot backend for that GUI would have to include
an implementation of those layout managers.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From kernr at mail.ncifcrf.gov  Fri Aug 20 12:53:30 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Fri, 20 Aug 1999 16:53:30 GMT
Subject: pythonwin
References: <37BCF0DD.EB99F9F0@netscape.com>
Message-ID: <37bd86f1.6269513@news.erols.com>

On Thu, 19 Aug 1999 23:08:29 -0700, vijay Mallela
<vijaym at netscape.com> wrote:

>does pythonwin has 1.5.2 v of python.. I am new to python and the
>download page says that includes build 125.. can some one clarify
>thx
>v

The Pythonwin distribution (IIRC) does not include the Python
interpreter.  You will need to download and install the Python 1.5.2
distribution first and then install Mark Hammond win32all extensions
(build 125 refers to these extensions, not the Python interpreter).
So download both py152.exe and win32all-125.exe.  I think that
Pythonwin is in win32all-125.exe.

Have fun.

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From wtanksle at dolphin.openprojects.net  Tue Aug 31 02:23:54 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Tue, 31 Aug 1999 06:23:54 GMT
Subject: running python on a floppy?
References: <z6ex3.1541$1i4.128395@typhoon.mbnet.mb.ca> <37CA50EC.4FC03560@home.com> <slrn7sm5m2.k0.cjc26@tankgrrl.bridget.mindriot.net>
Message-ID: <slrn7smt7p.geq.wtanksle@dolphin.openprojects.net>

On 31 Aug 1999 03:39:24 GMT, Cliff, or a close facsimile wrote:
>Pada Mon, 30 Aug 1999 09:32:42 GMT, Doug Hellmann bilang:
>| Why are people still trying to fit distributions like this on a
>| *floppy*?  Why not use a Zip disk?  Granted, they are more expensive,
>| but there is *so* much more room.

>After losing two zip disks within a month of each other to the Zip Click
>of Death, I vowed never to go near those things again. :P

I had understood that the click of death was a har4dware problem in the
earlier drives, not on the disks themselves.  At any rate, if you're
comparinng Zip disks to floppies, Zips win -- most floppies I've put
semi-important information on have failed, one just today barely half a
minute after I placed the info on it.

I've never had a Zip fail me, ever.

Of course, NOW I'm asking for it ;-).  I'd better make some new backups.

>cliff crawford   http://www.people.cornell.edu/pages/cjc26/

-- 
-William "Billy" Tanksley



From pehr at alum.mit.edu  Fri Aug  6 00:59:57 1999
From: pehr at alum.mit.edu (pehr anderson)
Date: Fri, 06 Aug 1999 00:59:57 -0400
Subject: downloading web based email
References: <7och4d$jr0$1@nnrp1.deja.com> <14249.55454.471036.787883@dolphin.mojam.com>
Message-ID: <37AA6BCD.1A611445@alum.mit.edu>

Yahoo *definitely* supports reading from their webmail accounts
though the universal standard of pop3.  Fetchmail will get this,
or you could simply point your netscape mail client at the yahoo server.
To enable this feature under yahoo, you have to click an option which
allows them to send you adds. Adds pay for their service so if you don't
go to their web page they have to be able to still get your attention.

I'm really impressed with how yahoo treats their clients.
Yahoo mail lets you specify the return address so you aren't
"locked in" to a yahoo account name, you can have mail forwarded from
a more permanent mail address and set your reply address to the permanent one!

    ---pehr


Skip Montanaro wrote:

>     Preston> I'm thinking of writing a Python program to download email from
>     Preston> web-based services such as Yahoo mail, Deja, Hotmail, and so
>     Preston> on, and deliever the mail in a format that standard unix
>     Preston> mailers can use.
>
>     Preston> I'm posting here to make sure that this work has not already
>     Preston> been done.  If someone has already seen a program like this,
>     Preston> that can be adapted for a variety of web based email services,
>     Preston> please let me know.  I'm not writing this as a learning
>     Preston> exercise, I just need the tool. ;-)
>
> Presuming the services you're interested in support a standard mail transfer
> protocol like POP or IMAP, you need look no further than Eric Raymond's
> excellent fetchmail program:
>
>     http://www.tuxedo.org/~esr/fetchmail/
>
> If not, perhaps you can train fetchmail to use whatever transport they do
> offer.
>
> Skip Montanaro  | http://www.mojam.com/
> skip at mojam.com  | http://www.musi-cal.com/~skip/
> 847-971-7098




From mal at lemburg.com  Wed Aug 18 06:25:41 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 18 Aug 1999 12:25:41 +0200
Subject: Class Description
References: <37ba0d28@news.interq.or.jp>
Message-ID: <37BA8A25.40C57268@lemburg.com>

Glen Malley wrote:
> 
> Obviously I'm a newbie at this stuff, so go easy on me.
> I'm just curious, having not read a great deal of the
> object-oriented gear in Python...
> Can a class print out info about itself?

No, Python doesn't have class methods. You can do the same
thing via special functions accessing the internal of classes
though. See hack.py on my Python Pages for some examples.

> For example, I know classes can be dynamically added to...
> ...could a class re-write itself, if properly messed with?

Yep. No problem at all. Some of this was just recently
introduced in 1.5.1 or 1.5.2 -- can't remember, e.g. you
can assign to __bases__, __dict__, etc. special class
attributes allowing you to change the behaviour *after*
creation.

See the docs for more infos, but beware: this is not really
newbie stuff, you're up to :-)

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   135 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From mcfletch at vrtelecom.com  Tue Aug 31 21:49:35 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Wed, 1 Sep 1999 01:49:35 GMT
Subject: lex / yacc for Python
In-Reply-To: <3d4shfiprl.fsf@amarok.cnri.reston.va.us>
References: <3d4shfiprl.fsf@amarok.cnri.reston.va.us>
Message-ID: <NDBBIAPHEKGOBIHCBLIIOEFPCAAA.mcfletch@vrtelecom.com>

Normally I point new users not to mcf.pars (which I haven't touched in ages
(though I still use it daily at work)) and point them to simpleparse, which
generates tuples for use with the mxTextTools engine.

Neither system is a lexx/yacc style system.  They use very similar general
pattern matching machinery which does not use lexxing machinery at all (i.e.
no tokenisation pass).  Both systems are targeted at (moderately) fast
parsing of a very limited class of grammars.

http://members.home.com/mcfletch/programming/simpleparse/simpleparse.html

As noted on your page, mcf.pars is still available as part of mcf.vrml, but
the version of mcf.vrml on starship.python.net is actually an old version of
that package.  I would have to check to see if the mcf.pars package changed
between that version and later versions of mcf.vrml.

Cheers,
Mike

-----Original Message-----
From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
Behalf Of Andrew M. Kuchling
Sent: August 31, 1999 5:50 PM
To: python-list at cwi.nl
Subject: Re: lex / yacc for Python


"Jacques Oosthuizen" <jacques at integra.co.za> writes:
> Any one know if there is such a beast

Quite a few, actually; many Python users seem to be language geeks.
A list of parsing systems is at:

http://starship.python.net/crew/amk/python/string.html#parsing

I've never used any of them, and don't know which ones are still
maintained, which are toys, and which are the most powerful.  Good luck!

--
A.M. Kuchling			http://starship.python.net/crew/amk/
This isn't life in the fast lane, it's life in the oncoming traffic.
    -- Terry Pratchett, in alt.fan.pratchett






From jstevens at basho.fc.hp.com  Fri Aug 13 22:53:51 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 19:53:51 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132253.QAA06419@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 4:53 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cc2f@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :Understanding somebody elses Perl. . . now that can be heart breakingly
> :difficult.
> 
> Then they were a bad Perl programmer, or else you are.

You are so big on assertion vs. proof: so prove the above.  If you
can.

> Perl doesn't force
> you to write good code any more than English forces to write good prose.
> When you find bad prose, whom do you shoot?  The writer, not the language.
> So, too, with Perl.  

Comparing Perl to English, and trying to emulate English, is in fact the
basis for Perl's flaws.

But the discussion was aimed more at "typesetting" not at "prose".

When you find bad typesetting, who do you shoot?  The system that
encourages/allows bad typesetting.  In this case, the flawed system
is Perl.

> If it is counted against Perl fault that "stupid" people can program in
> Perl, then so be it.

That is your mouth . . . this is mine.  I said nothing, ever, about
stupid programmers.

My criticism goes towards what should be obvious: slang can and does
make it more difficult to understand what others are saying, until
you learn that slang.  The more slang (the more variants allowed)
the more difficult a language is to learn.

Just ask any non-native speaker who is trying to learn American English.

> This aspect has made many people's lives better,
> even if they are too "stupid" or too "inexperienced" for a "real
> programming language".

Oh, Perl is as close to a real programming language as you can get. . .
assuming you subscribe to Porters definition of a "real programmer"! ;->

> They got their job done, and they're happy.

"They" may be happy.  Their team members are another story.

> I doubt you want to maintain their code, but why did you hire someone to
> write professionally maintainable code who had no demonstrated ability
> to do so?

In point of fact, when the projects mentioned before were shifted to
Python, three things happened:

1) Productivity went up.
2) Defects went down.
3) Maintenance costs went down.

Maybe it is Python that is easier for stupid programmer to use, not
Perl?

John S.




From andy at robanal.demon.co.uk  Fri Aug 13 16:55:48 1999
From: andy at robanal.demon.co.uk (Andy Robinson)
Date: Fri, 13 Aug 1999 20:55:48 GMT
Subject: UK mailing list
Message-ID: <37ba8623.4779222@news.demon.co.uk>

With help from Chris Tismer, there is now a Python UK mailing list.
You can sign up at:

http://starship.python.net/mailman/listinfo/python-uk

This is intended to be a low-volume list to help build a community in
the UK.  I hope we will be able to set up an informal user group,
exchange information on Python employment opportunities and
requirements, and work towards some Python events here.

Anyone on the continent or further afield who thinks they might like
to attend Python events in the UK is of course welcome.  

The acronym contest for the UK User Group(or groups)  is officially
declared open...

Regards,

Andy Robinson



From ScherBi at BAM.com  Tue Aug 24 12:09:42 1999
From: ScherBi at BAM.com (ScherBi at BAM.com)
Date: Tue, 24 Aug 1999 12:09:42 -0400
Subject: UML
Message-ID: <3F8BA119B48AD11197470008C7A493CB98B52F@HQOBGEX1>

>Thanks for your reply Bill.
You're quite welcome.

>I'll try it the other way around. Is the UML something like a OO
>Database or something like XML? Or anything in between?

For starters, the 'L' stands for language in both.  Let me quote from
_UML_In_A_Nutshell_ here:

"The UML is a modeling language for specifying, visualizing, constructing,
and documenting the artifacts of a system-intensive process"

Note the 'system-intensive process'; it's not directed solely at software
design, but that seems to be where it has had it's greatest impact.  Other
applications may be business process modeling, factory process modeling,
etc.

>1.) Object Primer(Scott W. Ambler SIGS Books/
>     Cambridge University Press, 1995)
>2.) UML in a Nutshell (OReilly)

>Which - or would you recommend both?

I've not seen _Object_Primer_, but I'm quite happy with the Nutshell book.
I've seen others I didn't like as much.


>I started documenting my work on AI in a simple word processor, then
>thinking of creating a database (to get hold of the code and my
>thoughts) and recently I heard something about CASE tools and UML.
>I'm really in great need of a tool which helps me to keep overview and
>organize my work.

A UML tool, I think, could only be a part of the solution here.  It's not
going to provide a database to hold code.  There are mechnisms for handling
documentation, however these bits of documentation are typically attached to
specific 'objects' within the tool.


>>I use UML, via ObjectDomain (www.objectdomain.com).
>>I am involved in developing ....
>>.. In other words, create classes and class diagrams in
>>UML, design the class interfaces, and play around with the design at this
>>level.  When you think you have it right, generate the stub code and fill
in
>>the methods.  You can fill in the methods without getting caught up in the
>>interface issues, which makes things much easier.

>Feels like UML is a sort of IDE??
No, but a CASE tool which uses UML as it's modeling language is a bit like
an IDE.  Again uml is just a language, like Python, C, or Latin.  

>>..where you understand the hierarchy and the operations, then
>>reverse engineer it into objectdomain.  You can then automatically
generate
>>a class diagram.  

>Something like a class browser used in C/C++ IDE's like Watcom?
Yes, I suppose.  But it shows you the class hierarchies in UML.

>>And no, I don't work for ObjectDomain.  It's the only UML tool I know of
the
>>generates and reverse engineers Python.  OD is written in Java.  I don't

>Whats so wrong about Java? When you use different OS's - Java should a
>logical way to go.

It's a bias born of experience ;-). The idea, the promise, the goal of Java
is noble and worthy. The reality of Java falls short.  I would prefer a
native code version.  It would stand alone( no jvm version dependencies),
and it would execute faster.  I suspect it would be more stable, also.  That
said, I'm quite happy with OD, but if they came out with a native code
version for a suitable platform, I would drop the Java version immediatly.

UML is a big thing.  I'm still a UML neophyte, but I'm getting use out of
it.  It took time, reading and experimenting, to get to this point, but it
has been worth it.  Not only have I begun to learn UML, but my OO
programming skills and understanding have improved too.  My company has
found enough value in it to send me out for training.  If you think UML
might work for you, I can't stress enough the value in downloading a tool,
like OD, and giving it a spin.


>Thanks for your patience
Your welcome, 

Bill Scherer



From david.c.hoos.sr at ada95.com  Fri Aug 13 21:39:47 1999
From: david.c.hoos.sr at ada95.com (David C. Hoos, Sr.)
Date: Fri, 13 Aug 1999 20:39:47 -0500
Subject: take the 20 second lang. survey
References: <37B4B3F0.77E6C59A@netplus.net>
Message-ID: <GN3t3.4218$Rh.194251@typ11.nn.bcandid.com>

Luke Loeffler <floods at netplus.net> wrote in message
news:37B4B3F0.77E6C59A at netplus.net...
> Take the 20-second language survey.  No personal information must be
> entered.  This is only a survey steming from my personal interest--it is
> not affiliated with any organization.
>
> http://www.recordingwebsite.com/language.html


I didn't answer the survey, because when it came to the questions
about OOL, the first internationally standardized OOL (i.e., Ada)
was not the choices offered.

With bias like that it's no wonder surveys seldom reflect the truth.






From mlh at idt.ntnu.no  Fri Aug 20 15:30:59 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 20 Aug 1999 21:30:59 +0200
Subject: Unsorting(randomizing) a sequence
References: <000701bee917$d8161cc0$eaa22299@tim> <14268.10663.137315.893101@buffalo.fnal.gov> <ZE1v3.14955$gO1.492954@news2.giganews.com> <37BCB288.B4476B8B@vision.arc.nasa.gov>
Message-ID: <y0jaerm45cc.fsf@vier.idi.ntnu.no>

Chad Netzer <chad at vision.arc.nasa.gov> writes:

> Christopher Browne wrote:
[...]
> The solution posted previously by Dan Schmidt, should work (I'll translate
> to C pseudocode to match the above form):
> 
> for (i = n; i > 0; i--)
> {
>   j = i -1;
>   swap(item[j], item[random(j)]); /* choose from remaining items in list */
> }

Which would be, in Python:

from random import randint

def shuffle(list):
    for x in range(len(list)-1,0,-1):
        y = randint(0,x-1)
        list[x],list[y] = list[y],list[x]

Is that among the suggested solutions? (Sorry for not really paying
attention... ;)

At least it's short...

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From godzilla at netmeg.net  Mon Aug  9 17:22:07 1999
From: godzilla at netmeg.net (Les Schaffer)
Date: Mon, 09 Aug 1999 21:22:07 GMT
Subject: looking for py/database ideas
Message-ID: <m2so5slkdj.fsf@netmeg.net>

a business/suit type friend of mine apprached me the other day about
helping him port an existing commercial Paradox database app to
something ""more modern"".

having vanishingly small database experience, i proceeded first to
browse the bookshelves for tips, and then after asking around some
amongst software friends, have learned that using MS Access / DAO /
VBA is considered de rigeur these days.

needless to say, this has induced derigor mortis in me. 

so i am fishing for ideas on other ways to build database
applications, and having just built a 10k line engineering simulation
app in python last year, i am interested in hearing if pythoners have
ideas for implementing comerical-type databases in python.

my buddy doesnt know what size market he has for this yet, but it will
initially be aimed at local municipalities in the US (small to medium
size towns and cities) who need to keep track of things (i shouldnt
say what things yet) which live in that municipality. as such, we are
not talking about high-availablity requirements -- wont be more than a
handful to tens of people (this is high) accessing the database at any
one time.

web access to db MIGHT be important if people in the field needed to
add data to the DB over a network connection (again, not a lot at one
time). 

the main concern over the existing Paradox DB is upgrade to what my
friend thinks is a more modern db system -- thats his opinion, i have
none since i know nothing about this -- whatever that means. i think
he thinks moving to a newer db system would allow for better solutions 
for db needs, and would move away from paradox which he thinks is a
dying db system. 

so..... i smell an opportunity to point him towards alternative
solutions. 

can things be done with python to create databases with a decent GUI
interface and Web access (err, not MS access) -- i am playing with
Zope a lil right now -- and SQL and all those goodie buzzwords. and
how hard would it be to integrate a python-esque type solution with
existing standard off-the-shelf systems running in other departments
(in other words, if another db system in that municipality used, say,
Acceess, could our solution be --- uhhh -- integrable with that system 
(boy do i need some buzzwords). 

just looking for some pointers to help my friend who, you may not
believe this from reading the above, is far more clueless than me.

:-)

thanks....

les schaffer



From skip at mojam.com  Sun Aug 29 22:54:30 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sun, 29 Aug 1999 21:54:30 -0500 (CDT)
Subject: Returning none
In-Reply-To: <37C7B86E.647BA654@prescod.net>
References: <37C55044.8CA02530@prescod.net>
	<37C5683D.91D61FD0@rubic.com>
	<14277.28442.44107.591412@dolphin.mojam.com>
	<37C70821.92E810FC@prescod.net>
	<14279.9024.401900.142461@dolphin.mojam.com>
	<37C7B86E.647BA654@prescod.net>
Message-ID: <14281.61572.191617.870559@dolphin.mojam.com>

    Paul> Makefile? Isn't that what I'm using Python to get away from? Even
    Paul> in a project with C components, I only run make when the C code
    Paul> changes, not when the Python code changes.

Using make is not synonymous with programming in C.  A Makefile is nothing
more than a record of dependencies between files and sets of actions that
are used to create the descendant files from their ancestors when they are
out of date.  If you have projects consisting of multiple files, you would
probably find make (or a clone) helpful.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From ebohlman at netcom.com  Fri Aug 13 22:55:19 1999
From: ebohlman at netcom.com (Eric Bohlman)
Date: 14 Aug 1999 02:55:19 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cf5f@cs.colorado.edu>
Message-ID: <7p2lqn$2pf@dfw-ixnews7.ix.netcom.com>

John W. Stevens (jstevens at basho.fc.hp.com) wrote:
[quoting someone whose name was lost]
: > It is wrong from the Larry's principle that things that behave
: > differently should have different appearances.
: 
: Then why doesn't Perl have two different operations for
: assign/reassign, vs. add to an array?  A tiny little violation of
: Larry's principles, that.

Nope.  "Behave differently" hear means "accomplish two different tasks."  
Comparing two strings to see if they're equal lexically is not the same 
task as comparing two strings to see if they're equal numerically.  
Putting a particular value in a particular position in an array that 
already holds a value *is* the same task as putting a particular value in 
a particular position of an array that does not yet hold a value; the 
only difference is in the internal details of how to accomplish it.

Of course, the OOically-correct way to handle the string comparisons is 
to define a base class, ComparableString, and two subclasses, 
LexicallyComparableSting and NumericallyComparableString, along with 
polymorphic operators that understand which class they're comparing to 
and apply the correct comparison method.  But there are cases where you 
may want to do *both* comparisons on a string, and to be OC you then have 
to cast the string to the appropriate type; the Perlish way of picking 
the operator director is IMHO easier and more maintainable; in my 
experience, heavy casting makes code unreadable.




From kernr at mail.ncifcrf.gov  Thu Aug 19 17:21:41 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Thu, 19 Aug 1999 21:21:41 GMT
Subject: More random python observations from a perl programmer
References: <199908191819.MAA04074@jhereg.perl.com>
Message-ID: <37bc71cc.19700984@news.erols.com>

On Thu, 19 Aug 1999 12:19:00 -0600, Tom Christiansen
<tchrist at jhereg.perl.com> wrote:

>>Yup.  It says that Windows help is useful for more people than icky
>>manpages.  
>
>No, it means that it's considered ok to alienate people who expect manpages.

Not really.  I have never seen anyone express a desire for manpages
before today.  There was a large discussion some time back on Python's
documentation.  While I didn't read every post of that discussion, I
don't recall anyone requesting manpages.  Many people were requesting
something like pod, though.  :-)

The reason we have Windows help files for Python's documentation is
because someone in the Python community (not Fred Drake, the official
documentation maintainer) wanted them and did the necessary work to
provide them.  The reason we don't have UNIX manpages is because not
enough people (if any besides you) have asked for them or done the
work.  Fred is always willing to accept volunteers, though.

In the meantime, python(1), which should be installed automatically by
the Makefile (if it didn't, the file "python.man" is in the Misc/
subdirectory of the source package), will refer you to the Python
Tutorial, the Python Library Reference, and the Python Reference
Manual.  Unfortunately, what it doesn't mention is the URL for getting
this documentation.  This is a holdover from the time before the
source and documentation were split into separate distributions.  If
nothing else is done about manpages, this oversight should be fixed.
Fred?

>--tom

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From arcege at shore.net  Mon Aug 16 20:53:10 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 17 Aug 1999 00:53:10 GMT
Subject: function which knows who has called it.
References: <934841569snz@vision25.demon.co.uk>
Message-ID: <Wn2u3.317$mV.112571@news.shore.net>

Phil Hunt <philh at vision25.demon.co.uk> wrote:

: Is if possible in Python to have a function that prints the name
: of the function that called it. (I'm thinking of using this in
: some debugging code I want to write).

The information is captured in the frames in the traceback.

  def you_rang():
    import sys
    try:
      raise RuntimeError
    except RuntimeError:
      exc, val, tb = sys.exc_info()
      frame = tb.tb_frame.f_back
      del exc, val, tb
    try:
      return frame.f_back.f_code.co_name
    except AttributeError:  # called from the top
      return None

  def f():
    WhoCalledMe = you_rang()

  -Arcege





From tchrist at mox.perl.com  Thu Aug 19 15:43:39 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 12:43:39 -0700
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7ro9ns.h6v.jblaine@shell2.shore.net> <37bc2ee5@cs.colorado.edu> <nJUYxAAwWEv3EwY0@jessikat.demon.co.uk>
Message-ID: <37bc505b@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Robin Becker <robin at jessikat.demon.co.uk> writes:
:It's a good thing I'm in the killfile as otherwise my response would be
:raising the blood pressure of this arrogant person. Perl with any amount
:of documentation sucks when compared to python. It's not so much the bad
:language structure/syntax but the lack of sensible advice. Also (it's a
:number thing) there are good reasons for maths/engineering types like
:myself to prefer python. 

I never killfiled you.  Why are you trying to start a flame war?  And why
did you respond with a stupid Jeopardy style Answer before the Question,
without even trimming?  Is your newsreader broken, or haven't you ever
bothered to read news.announce.newusers to learn how to use Usenet?

Followups to /dev/null.

--tom
-- 
The very fact that it's possible to write messy programs in Perl is also
what makes it possible to write programs that are cleaner in Perl than
they could ever be in a language that attempts to enforce cleanliness.
        --Larry Wall



From Dan at Grassi.com  Sat Aug 14 15:14:11 1999
From: Dan at Grassi.com (Dan Grassi)
Date: Sat, 14 Aug 1999 15:14:11 -0400
Subject: Mac ODBC
Message-ID: <7p4f3r$16t2@dilbert.whoi.edu>

Does anyone have ODBC working under Mac Python?  If so could you provide me
with some hints on where to get the various Mac sources and some compile
hints?  I do have a PPC Mac ODBC driver that does work.

Thanks,

Dan, Dan at Grassi.com



From henry299 at my-deja.com  Wed Aug 25 23:21:37 1999
From: henry299 at my-deja.com (henry299 at my-deja.com)
Date: Thu, 26 Aug 1999 03:21:37 GMT
Subject: MSDOS window
Message-ID: <7q2brn$u7h$1@nnrp1.deja.com>

Hi, i'm new at this. I just downloaded version
1.5.2 for windows 95. When i double click on a
program in the windows explorer, (example.py
which is  print "This is some example text") an
MS-DOS window opens and the text is displayed,
but the window closes by itself in less than a
second. Am i doing something wrong, and is there
a way to keep the popup window open? Any help
would be greatly appreciated.

Thanks in advance,

Matt


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From S.I.Reynolds at cs.bham.ac.uk  Wed Aug 18 07:06:51 1999
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Wed, 18 Aug 1999 12:06:51 +0100
Subject: Deepcopying numeric array
Message-ID: <37BA93CB.5050@cs.bham.ac.uk>

Does anyone know why the Numeric arrays can't be deepcopied? Is it just
an omission?

Cheers

Stuart


>>> from Numeric import *
>>> a = array([1,2,3])
>>> a
array([1, 2, 3])


>>> from copy import *
>>> deepcopy(a)
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/bham/ums/common/pd/packages/Python/lib/python1.5/copy.py", line
147, in deepcopy
    raise error, \
copy.error: un-deep-copyable object of type <type 'array'>
>>>



From thantos at brimstone.mecha  Fri Aug 20 03:50:02 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Fri, 20 Aug 1999 07:50:02 GMT
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu>  <14268.23429.917082.290102@anthem.cnri.reston.va.us>
Message-ID: <slrn7rq1ur.e45.thantos@brimstone.mecha>

On Thu, 19 Aug 1999 15:31:17 -0400 (EDT), Barry A. Warsaw
<bwarsaw at cnri.reston.va.us> wrote:

>It says that Windows users were more motivated to provide platform
>specific documentation than any Unix user so far.

More specifically, it says we Real Unix Men(tm) don't mind using the
HTML indices to the module docs on the Python site since Lynx works
just fine from our Wyse or VT100 character terminals ...  Why reinvent
the wheel?

Besides, language internals shouldn't be documented in seperate man
pages, that way lies madness in any language with any kind of complex
syntax.  Why reinvent the wheel?

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From hei at adtranzsig.de  Wed Aug  4 03:05:38 1999
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Wed, 4 Aug 1999 09:05:38 +0200
Subject: building embedded classes as in C/C++
References: <37A78579.9856D16@psych.ucsb.edu>
Message-ID: <933752678.796042@news.adtranzsig.de>

Python has no member variable declarations.
You create your member variables in a constructor.
Try this (didn't try it, hope there's no typo):

class Point:
  def __init__(self):
    self.x = 0

class Object:
  def __init__(self):
    self.pos = Point()
    self.ang = Point()
obj1 = Object()
obj2 = Object()

obj1.ang.x = 123

print obj2.ang.x

--
Dirk
Where did you want to go again?
Andy Beall schrieb in Nachricht <37A78579.9856D16 at psych.ucsb.edu>...
>Hi,
>
>I'd like to build an embedded structure (Python class) much like I do in
>C if possible.  When I do the following, multiple instances of my final
>class overwrite each others sub-class values!
>
>-------------------------------------------------------------------------
>class Point:
> x = 0
> y = 0
> z = 0
>
>class Object:
> pos = Point()
> ang = Point()
>
>
>obj1 = Object()
>obj2 = Object()
>
>obj1.ang.x = 123
>
>print obj2.ang.x    <<-- AND I FIND IT CONTAINS OBJ1's X VALUE
>-------------------------------------------------------------------------
>
>
>
>In C, this would be:
>
>-------------------------------------------------------------------------
>typedef struct Point {
> float  x;
> float  y;
> float  z;
>} Point;
>
>typedef struct Object {
> Point  pos;
> Point  ang;
>} Object;
>
>Object obj1;
>Object obj2;
>
>obj1.ang.x = 123;
>printf("%f", obj2.ang.x);   <<-- AND NOW THE VALUES IN OBJ1 AND OBJ2
> WILL NOT BE RELATED (OBJ2 WILL JUST
> CONTAIN GARGAGE AT THIS POINT).
>-------------------------------------------------------------------------
>
>
>I've found that writing something like:
>
>class Object(Point):
> ...
>
>works, but then I don't have a wrapper around the (x, y, z) values
>anymore and I
>don't know how to bring in Point more than once. Is it possible to do
>this in Python.  I've looked in the tutorial and Mark Lutz's book to no
>avail.
>
>Thanks,
>Andy Beall





From Ted.Horst at wdr.com  Mon Aug 30 19:38:26 1999
From: Ted.Horst at wdr.com (Ted Horst)
Date: Mon, 30 Aug 99 18:38:26 -0500
Subject: threads on Solaris
Message-ID: <199908302338.AA04307@ch1d2833nwk>

Does anyone know why I get 4 threads when I start the interpreter under  
Solaris (as reported by /bin/ps -aL) ?

This is python 1.5.2 built with a SUNPro compiler (sorry, I don't have the  
config options).

> uname -a
SunOS ch1d2964syb 5.6 Generic_105181-12 sun4u sparc SUNW,Ultra-Enterprise


Thanks,

Ted


---



From chinaski_py at my-deja.com  Mon Aug 23 18:23:34 1999
From: chinaski_py at my-deja.com (chinaski_py at my-deja.com)
Date: Mon, 23 Aug 1999 22:23:34 GMT
Subject: NEWBIE:Let's try that again..Help w/path
References: <7pm0o4$e0h$1@nnrp1.deja.com>
Message-ID: <7pshkp$q47$1@nnrp1.deja.com>

In article <7pm0o4$e0h$1 at nnrp1.deja.com>,
  chinaski_py at my-deja.com wrote:
> I'm new and need help with my Python path.When I go to open a Python
> file (.py) it won't open .I go to (on Win98):
> start
> run
> sysedit
> and here is my path:
> path %path%;C:\Perl\bin;path=C:\PROGRA~1\Python\;%path%
>                 Any help on this would be appreciated,
>                                 G.Norton (chinaski_py)
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>
Thomas,
      tried your path but can't open "IDLE".The icon is there in the
file, but I click and nothing happens.This is what I get:
                       PROGRAM NOT FOUND
                 This program is needed for opening files of type
                     'Python File' (no console)
shouldn't it be in the path? Glenn (chinaski)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From skip at mojam.com  Wed Aug 25 11:20:15 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 25 Aug 1999 10:20:15 -0500 (CDT)
Subject: Small question about Python executeable(sp?) files
In-Reply-To: <37C3DAF9.D51DAAF5@uswest.net>
References: <37C3DAF9.D51DAAF5@uswest.net>
Message-ID: <14276.2373.676736.425421@dolphin.mojam.com>

    Syniq> I know that
    Syniq> #! /something/blah/whatever/python

    Syniq> as the first line in your code will make your "program" an
    Syniq> executeable(sp?), but this assumes that your user has python
    Syniq> installed in /something/blah/whatever/python.

    Syniq> What I'm wondering: Is there a way to make the script execute
    Syniq> regardless of the location (or even existence) of python on the
    Syniq> user's machine (basically, compile it into something sort of
    Syniq> resembling ye old .EXE file)?

Two things come to mind.  One, assuming Python is somewhere on the user's
path, use

    #!/usr/bin/env python

as the first line.  Two, check out the various freeze/squeeze/xfreeze
packages available for Python.  Poking around http://www.python.org/search
should yield some positive leads.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From info at pythonware.com  Fri Aug  6 05:17:24 1999
From: info at pythonware.com (PythonWare)
Date: Fri, 6 Aug 1999 09:17:24 GMT
Subject: ANN: PIL 1.0 final
Message-ID: <008e01bedfec$7ffe7110$f29b12c2@secret.pythonware.com>

The Python Imaging Library (PIL) adds image processing capabilities
to your Python interpreter. This library supports many file formats,
and provides powerful image processing and graphics capabilities.

PIL 1.0 final is now available from:

    http://www.pythonware.com/

Changes since the last beta include faster text rendering, support
for using PIL as a package, mechanisms to hook into a standard Tkinter
build, better GIF encoding, incremental image decoding, and various
bug fixes.  see the README and CHANGES files for the full story.

PIL 1.0 is (as before) shipped under a Python-style license, and can be
used without fee also in commercial projects.  However, we encourage
commercial users to support our development efforts by purchasing a
support package; for details, see:

    http://www.pythonware.com/products/pil/support.htm

enjoy,
the PIL development team





From info at pythonware.com  Sat Aug  7 10:35:20 1999
From: info at pythonware.com (PythonWare)
Date: Sat, 7 Aug 1999 14:35:20 GMT
Subject: ANN: PIL 1.0 final
References: <008e01bedfec$7ffe7110$f29b12c2@secret.pythonware.com> <37aadb42.108851580@news.u-psud.fr> <m37ln9i4or.fsf@schwinger.harvard.edu>
Message-ID: <018001bee0e2$14502520$f29b12c2@secret.pythonware.com>

Michael Spalinski <mspal at sangria.harvard.edu> wrote:
> I compiled PIL 1.0 on Red Hat Linux 5.2, but then I get:
> 
> [schwinger:/usr/src/python/Python-1.5.2/Extensions/Imaging-1.0]# python
> Python 1.5.2 (#5, Apr 14 1999, 19:21:58)  [GCC 2.7.2.3] on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import _imaging
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> ImportError: /usr/lib/libtk8.0.so: undefined symbol: XFreePixmap
> >>>   
> 
> Is my tk version too old? I have tk-8.0.3-20.

this might work:

    comment out the Tkinter-related lines in the
    Setup file and rebuild PIL (to use ImageTk, you
    need to hook into Tkinter in the old way; see
    Tk/README for more info).

or this:

    rebuild Tcl/Tk with --enable-shared, and then
    relink PIL.

hope this helps!

the PIL development team




From esuzm at primrose.csv.warwick.ac.uk  Sun Aug 15 17:56:18 1999
From: esuzm at primrose.csv.warwick.ac.uk (Andrew Clover)
Date: 15 Aug 1999 21:56:18 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <37b5a8e0@cs.colorado.edu>
Message-ID: <7p7d22$1hh$1@holly.csv.warwick.ac.uk>

Tom Christiansen (tchrist at mox.perl.com) wrote:

> Common rebuttals [to += and co.] include:

>     3) It's more efficient to keep a tmpobj anyway.

Bah. Temporary variables are evil!

>     4) Redundancy can clarify through redundancy.

Wossat? Sounds like rot to me. :-)

> As to why there's no x++ there, consider how often people from a
> Pascal(ish) background are confused between x+1 and x++ and ++x, and
> you'll begin to see what kind of problems caused it to be vetoed.

This is a good point, but stems from C's tendency to return values all
the time from things I think it shouldn't. :-)

I'd like a syntax like:

  x.+(1)

With the idea that x is an object (of type int) with a '+' method that
adds stuff to it. It'd be nicer without the brackets though.

:opens the language can of worms again

-- 
This posting was brought to you by And Clover.
(Sorry.)

http://radio.warwick.ac.uk/shows/losertown/



From fredrik at pythonware.com  Thu Aug 26 06:28:52 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 26 Aug 1999 10:28:52 GMT
Subject: anyon know anythin about "squeeze"?
References: <s7c26f14.054@mail.conservation.state.mo.us>
Message-ID: <00b301beefad$ce7f6a30$f29b12c2@secret.pythonware.com>

Jeffrey Kunce <kuncej at mail.conservation.state.mo.us> wrote:
> I see freeze, but not 'squeeze' in my 1.5.2 (I don't have CVS access).
> 
> The copy of squeeze that I use came from Fredrik Lundh at:
> http://starship.skyport.net/crew/fredrik/ipa/squeeze.htm 

the current version is available at:
http://www.pythonware.com/downloads.htm

</F>




From tchrist at mox.perl.com  Thu Aug 19 15:33:09 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 12:33:09 -0700
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <37bc2ee5@cs.colorado.edu> <7phfe0$19so$1@nntp6.u.washington.edu> <37bc487e@cs.colorado.edu>
Message-ID: <37bc4de5@cs.colorado.edu>

I wrote:
:I agree entirely, and Perl shouldn't be doing that.  It uses 
:it's only manpath, which is separate.  For example:

I surely meant

    its own manpath

not 

    it's only manpath

--tom
-- 
#define SVs_RMG   0x00008000  /* has random magical methods */
    --Larry Wall, from sv.h in the v5.0 perl distribution



From thstr at serop.abb.se  Thu Aug 12 08:28:00 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Thu, 12 Aug 1999 14:28:00 +0200
Subject: internet programming with python
References: <37AF678E.CC8E165B@netscape.com>
Message-ID: <37B2BDD0.F86BC914@serop.abb.se>

Hi
 
Here is a Swedish site that has it: 
   www.bokus.com
Press the "sverige" button for next page
In the little box to the left labeled "snabbs?k"
write "python" hit enter. And then you get a list of books.

I guess to order you might wanna send a mail to 
   info at bokus.com
to get details about how to order outside Sweden.
   Internet Programming With Python
   ISBN: 1558514848
   Authors: Watters, Aaron
            Rossum, Guido Van
            Ahlstrom, James C.

It also seems to exist at Amazon.de; follow the link from 
 http://www.python.org/doc/Books.html
where other books may be found as well.

I have read some of "Programming Python" by Mark Lutz and found it
quite helpful. 

Good Luck

  -- Thomas

Vijay Mallela wrote:
> 
> Hi,
> I am looking for the book "Internet programming with python". I need it
> at the earliest..
> I checked out amazon and other stores and they dont have it.. does any
> one know where I can get the book
> Thanks
> vijay



From sasha at ofoto.com  Mon Aug 23 13:09:33 1999
From: sasha at ofoto.com (Alexander Voynow)
Date: Mon, 23 Aug 1999 10:09:33 -0700
Subject: PyApache / httpdapy on NT
Message-ID: <sefw3.928$3U2.2145@dfw-read.news.verio.net>

<apologies for possible duplicate post>

Hi all,

 I've got a python cgi app that's currently running on Solaris
 with PyApache for a bit of a  performance boost.
 It looks as if it may have to be moved over to an NT platform.
 Before I start barking up the wrong tree, is it possible to compile
 PyApache or httpdapy for Win32? I will need one or the other as
 I don't think I can afford the cgi overhead.
 Caveat: VC++ is fairly foreign territory to me, so be gentle.

 Of course, if anyone has binaries.....  :  )

 Sasha
sasha at ofoto.com





From quinn at cruzeiro.ugcs.caltech.edu  Sat Aug 21 23:07:14 1999
From: quinn at cruzeiro.ugcs.caltech.edu (Quinn Dunkan)
Date: 22 Aug 1999 03:07:14 GMT
Subject: What is a troll by the way? [offtopic]
References: <zelv3.13444$x04.816752@typ11.nn.bcandid.com> <37bdeb1b.31915768@news.erols.com> <37BDF3A7.75FEA04F@vision.arc.nasa.gov> <Z1xv3.11$EN1.7779@news.shore.net>
Message-ID: <slrn7ruqb8.sjv.quinn@cruzeiro.ugcs.caltech.edu>

On Sat, 21 Aug 1999 12:35:05 GMT, Michael P. Reilly <arcege at shore.net> wrote:
>Chad Netzer <chad at vision.arc.nasa.gov> wrote:
>: Robert Kern wrote:
>
>:> _Troll_ is a USENET term for someone that deliberately posts an
>:> inflammatory message simply to stir up argument.
>
>: I've found it funny to watch this term evolve over the past few years.
>: It started (I believe) as a reference to the fishing term, and used to
>: mean "trolling for flames" (ie. hoping to bring flames to the
>: discussion), perhaps employing "flame bait" for this purpose.

You've just learned another usenet principle: threads veering off-track into
detailed obscurity as people feel obliged to reveal their knowledge.  You can
learn a lot of interestingrandomuseless this way.

>It is the same with me; I was around during the Spam Wars, when the
>term "spam" was first created (yes, it was after the Monty Python
>sketch).

Hmm, even if it's not the original reason the term came into use, spam still
carries the nice image:

"But I don't want ANY spam!"
"... well, you could have spam, bacon, and eggs.  That's not got MUCH spam in
it."

No matter how much you try to get rid of it, it's still there.



From tim_one at email.msn.com  Sat Aug 14 23:09:54 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 14 Aug 1999 23:09:54 -0400
Subject: Regarding the magical "or" operator
In-Reply-To: <37b5d712@cs.colorado.edu>
Message-ID: <000601bee6cb$a5e747e0$832d2399@tim>

[posted & mailed]

[Tom Christiansen]
> I was wondering how often most python programmers make use of that
> remarkably nifty property of the logical operators, in which they
> produce not merely 0 or 1, but the value is last looked at when it
> decided whether the expression was true or false.

Hard to say.  It shows up dozens of times in the std libraries, but like
other features that aren't exactly like C's, I'd guess "most programmers"
don't think to use it ("else" clauses on loops are similar this way).

The most common idiom is

    value_to_use = some_value or some_default_value

A cute one is in the Wichmann-Hill random # generator, where a seed int
cannot be allowed to become 0:

	x = (x + a) % 256 or 1

> ...
> My question is this: have you ever had an urge for it (or a variant)
> to care about the difference between None and any other false value?

Whaddya think this is, Perl <wink>?

> I am certainly *not* asking for such a thing.  I'd just like to know
> whether anyone ever cared.

Not that I recall.  The notion that "one distinguished false value is more
false than the others" wouldn't fly well here.  None doesn't come with
enough semantics (either enforced or conventional) to support making subtle
distinctions -- None evaluates false, and has an easy-to-remember name, but
there's nothing else special about it in theory or practice.

It may be different in Perl, where e.g. I imagine that an operator returning
the first non-undef value could be handy.  It would be handy in Python too
*if* e.g. unbound names were instead bound to None by default (instead of
raising exceptions upon reference).

In Python-land folks are more likely to clamor for an equivalent to Perl/C

    a = b() ? c() : d()

The simplest faithful Python one-liner

    # trickiness needed in case c() evaluates false;
    # [c()] is still true in that case
    a = (b() and [c()] or [d()])[0]

makes

    if b():
        a = c()
    else:
        a = d()

look downright appealing <wink>.

none-is-its-own-reward-ly y'rs  - tim





From mhammond at skippinet.com.au  Wed Aug  4 21:07:17 1999
From: mhammond at skippinet.com.au (Mark Hammond)
Date: Thu, 5 Aug 1999 11:07:17 +1000
Subject: Printing from python on Windows platform
In-Reply-To: <14248.57695.330650.399604@dolphin.mojam.com>
Message-ID: <006401bedede$dde85750$1101a8c0@bobcat>

> Dumb question from a non-Windows person: shouldn't that flag
> arg be "wb"?
> We Unix weenies are always getting beat up by the Windows
> weenies for not
> including the "b"inary flag when opening files.  I figured
> turn about is
> fair play. ;-)

I have no idea at all :-)  We are talking to the printer as a tty device,
so I guess it depends on the printer.

So I guess the "portable" thing do is indeed to open in binary mode, and
send explicit line terminators of your own choosing.  But how do you know
the correct terminator for the printer?  '\r\n' is probably safest for a
printer in TTY mode.

Specifying text mode make some sense on Windows at least - you can send
normal "\n" characters, and they will be translated to "\r\n", which is
likely to be what you want.

[But your original point is taken :-]

Mark.




From sholden at pgrad.cs.usyd.edu.au  Sat Aug 14 22:12:27 1999
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 15 Aug 1999 02:12:27 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com>  <slrn7rb1uj.cf9.jstevens@bamboo.verinet.com>
Message-ID: <slrn7rc8h8.rsf.sholden@pgrad.cs.usyd.edu.au>

On Sat, 14 Aug 1999 15:14:28 GMT,
    John Stevens <jstevens at bamboo.verinet.com> wrote:
>On Sat, 14 Aug 1999 12:06:11 +0100, Alex Maranda <amaranda at nospam.com> wrote:
>>Sam Holden wrote: 
>>Why? You don't happen to hate Pythoneers or Python itself, do you? I've
>>actually appreciated the argument from your second post, when you taxed
>>the other guy for needlessly questioning your Python knowledge out of
>>context.
>
>It was not out of context.
>
>Look at the Subject of this thread. . .

Read the content of my post, and learn how usenet wanders... I wasn't
arguing about the relative merits of perl and python. I don't care, I 
use both, hang I've used them both in the same system.

-- 
Sam

compiling kernels is what I do most, so they do tend to stick to the
cache ;)	--Linus Torvalds



From p.t at iname.com  Sat Aug  7 06:08:04 1999
From: p.t at iname.com (Peter Torstenson)
Date: Sat, 7 Aug 1999 07:08:04 -0300
Subject: How to use KConfig - pyKDE 0.8?
Message-ID: <7oh0gm$1sio@enews3.newsguy.com>

I'm trying to find out how to read and write to the KDE application rc file.
Since I'm using KTMainWindow a configurationfile is created.
Now I would like to add my own parameters to this file so that I don't have
to create another config-file.
I found som info on KConfig on the KDE site, but I can't get it to work.
Is there anyone that kan supply an example of how to read and write to a KDE
application rc file?

Greateful for help!

/Peter Torstensson






From JamesL at Lugoj.Com  Mon Aug 16 23:33:01 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Mon, 16 Aug 1999 20:33:01 -0700
Subject: How to write to a named pipe?
References: <7p87nv$fbl$1@nnrp1.deja.com>
Message-ID: <37B8D7ED.4A4D5CCC@Lugoj.Com>

singlets at my-deja.com wrote:
> >>> import os
> >>> os.mkfifo('/tmp/myfifo')
> >>> f=open('/tmp/myfifo','w')    # now we're stuck

Try:

f = open('/tmp/myfifo', 'w+')

(Note the extra '+')

Not sure why 'rw' doesn't work. But 'r+' also works. Here is something you
can try: in one window start a Python interpreter and do:

f = open('/tmp/myfifo', 'w')   # The usual block.

And then in the other interpreter do:

f = open('/tmp/myfifo', 'r')    # Both should now return!

You can reverse the open order and get the same result. Basically pipes are
unidirectional and expect a reader and a writer. I'm so used to opening
pipes in 'C' in read/write mode that I've forgotten why things are like
this. At least it doesn't block, once you remember to always open read/write
or similar.



From egelando at online.no  Thu Aug  5 06:45:16 1999
From: egelando at online.no (Egeland)
Date: Thu, 5 Aug 1999 12:45:16 +0200
Subject: How to link statically with python15.lib....
Message-ID: <sYdq3.157$E%3.952@news1.online.no>

Hello. Could someone please tell me how to link statically to python15.lib
using MSVC6.0 Prof. Edit. ?
I want to do this in order to avoid messing with *.dll's that may reside on
the installation platform. I'm not
really into install shield or anything ; just wanted to do it the easy way
(if there is one). Guess I'll have to
link to tcl80.lib and tk80.lib as well..?

Thanks in advance,

Anders Moe





From mal at lemburg.com  Tue Aug  3 15:04:30 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Tue, 03 Aug 1999 21:04:30 +0200
Subject: Python BSCW Server Error AttributeError: altzone
References: <Pine.LNX.4.10.9908031250400.31183-100000@snet.wit.ie>
Message-ID: <37A73D3E.17812EBB@lemburg.com>

Kieran O' Sullivan wrote:
> 
> I am running BSCW on SuSE Linux 5.3 with Apache 1.3.  The BSCW will
> allow logins and creation of users but when ever a user whether they
> are an Admin or just na ordinary user tries to create a folder the
> following happens.
> 
> Unanticipated Error (BSCW Release 990623-1553):
> Traceback (innermost last):
>   File "wrapped", line 16, in ?
>     cl_request.run('bs_cgi')
>   File "BSCW3/pyc15/cl_request.py", line 1351, in run
>   File "BSCW3/pyc15/cl_request.py", line 801, in handle_rq
>   File "BSCW3/pyc15/cl_request.py", line 891, in handle_it
>   File "BSCW3/pyc15/op_get.py", line 65, in handle_GET
>   File "BSCW3/pyc15/cl_core.py", line 1166, in get
>   File "BSCW3/pyc15/bs_container.py", line 150, in get
>   File "BSCW3/pyc15/bs_container.py", line 274, in visible_entries
>   File "BSCW3/pyc15/bs_container.py", line 274, in <lambda>
>   File "BSCW3/pyc15/cl_core.py", line 98, in get_as_entry
>   File "BSCW3/pyc15/cl_core.py", line 946, in get_as_entry
>   File "BSCW3/pyc15/bs_artifact.py", line 261, in get_as_entry
>   File "BSCW3/pyc15/bs_artifact.py", line 91, in get_entry_params
>   File "BSCW3/pyc15/bs_interface.py", line 438, in event_msg
>   File "BSCW3/pyc15/bs_utils.py", line 304, in relative_time
>   File "BSCW3/pyc15/cl_request.py", line 295, in localtime
>   File "BSCW3/pyc15/cl_request.py", line 914, in __getattr__
>   File "BSCW3/pyc15/cl_request.py", line 1200, in def_timezone
> 
> AttributeError: altzone

This is a (known ?) error in the 1.5.2 release of Python. It only
shows up on libc5 systems. Here is the patch:

--- /home/lemburg/orig/Python-1.5.2/Modules/timemodule.c        Mon Apr  5 23:54:14 1999
+++ Modules/timemodule.c        Mon Apr 26 21:03:02 1999
@@ -614,11 +614,11 @@ inittime()
        p = getenv("PYTHONY2K");
        ins(d, "accept2dyear", PyInt_FromLong((long) (!p || !*p)));
        /* Squirrel away the module's dictionary for the y2k check */
        Py_INCREF(d);
        moddict = d;
-#if defined(HAVE_TZNAME) && !defined(__GNU_LIBRARY__)
+#if defined(HAVE_TZNAME) && !defined(__GLIBC__)
        tzset();
 #ifdef PYOS_OS2
        ins(d, "timezone", PyInt_FromLong((long)_timezone));
 #else /* !PYOS_OS2 */
        ins(d, "timezone", PyInt_FromLong((long)timezone));
@@ -632,11 +632,11 @@ inittime()
        ins(d, "altzone", PyInt_FromLong((long)timezone-3600));
 #endif /* PYOS_OS2 */
 #endif
        ins(d, "daylight", PyInt_FromLong((long)daylight));
        ins(d, "tzname", Py_BuildValue("(zz)", tzname[0], tzname[1]));
-#else /* !HAVE_TZNAME || __GNU_LIBRARY__ */
+#else /* !HAVE_TZNAME || __GLIBC__ */
 #ifdef HAVE_TM_ZONE
        {
 #define YEAR ((time_t)((365 * 24 + 6) * 3600))
                time_t t;
                struct tm *p;
@@ -681,11 +681,11 @@ inittime()
        ins(d, "altzone", PyInt_FromLong(timezone));
        ins(d, "daylight", PyInt_FromLong((long)0));
        ins(d, "tzname", Py_BuildValue("(zz)", "", ""));
 #endif /* macintosh */
 #endif /* HAVE_TM_ZONE */
-#endif /* !HAVE_TZNAME || __GNU_LIBRARY__ */
+#endif /* !HAVE_TZNAME || __GLIBC__ */
        if (PyErr_Occurred())
                Py_FatalError("Can't initialize time module");
 }
 
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   150 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From dfan at harmonixmusic.com  Thu Aug 19 12:39:40 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 19 Aug 1999 12:39:40 -0400
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7ro9ns.h6v.jblaine@shell2.shore.net>
Message-ID: <wklnb7k9mb.fsf@turangalila.harmonixmusic.com>

jblaine at shell2.shore.net (Jeff Blaine) writes:

| I don't know, your little jab there in an otherwise great reference
| document posting seemed completely petty to me.  It makes me think
| you stuck it in there as a little "See, I'm dissing them like I am
| supposed to!" due to the fact that you posted the message to both
| c.l.perl and c.l.python.

Actually, he left that paragraph out of the copy he posted to
comp.lang.perl.misc.

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From stephan at pcrm.win.tue.nl  Fri Aug 27 03:34:20 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 27 Aug 1999 09:34:20 +0200
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <slrn7sasvb.397.kzaragoza@kzaragoza.ne.mediaone.net> <slrn7scf7d.pa.clgonsal@keeshah.penguinpowered.com>
Message-ID: <m3wvuh8yo3.fsf@pcrm.win.tue.nl>

clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves) writes:

> It is possible, but more complicated, to make this a compile time error.
> This would essentially require a function/procedure distinction with
> restrictions about how you used return in each one. There would still
> have to be a runtime error for cases where someone tried to get a result
> from a procedure, but cases like the above example could be caught at
> compile-time. (The above, being a function, must return a value.
> Procedures would have to defined in some other way, like using some new
> keyword "proc" in place of "def")

Moreover, this would require different syntax when *calling* the 
function/procedure. Let's say, procedure names start with `&' ?

<wink>

Greetings,

Stephan



From skip at mojam.com  Sun Aug 29 23:02:01 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sun, 29 Aug 1999 22:02:01 -0500 (CDT)
Subject: echoing input
In-Reply-To: <19990829092044.A5809@optiplex.palga.uucp>
References: <7q92j8$naa$1@demon.uunet.ca>
	<19990829092044.A5809@optiplex.palga.uucp>
Message-ID: <14281.62365.82705.799831@dolphin.mojam.com>

    >> Hi.  Is there a way to get Python to echo the contents of a script
    >> while it is being loaded?  I have a script called x.py that contains:

    Gerrit> #!/usr/bin/python
    Gerrit> """
    Gerrit> echo the contents of a script
    Gerrit> """

    Gerrit> import sys

    Gerrit> SCRIPT = sys.argv[0]	# no, it's something else... (find the full path of the
    Gerrit> 			# script

    Gerrit> SCRIPT_CONTENTS = open(SCRIPT).read()

    Gerrit> print SCRIPT_CONTENTS

    Gerrit> # EOF

I don't see the script being executed anywhere.  I think what the original
poster asked for was a mode similar to the Bourne shell's -x flag.

If you don't might a horrendous impact on performance, look at my trace
module (in http://www.musi-cal.com/~skip/python/).  If you set verbose mode, 
it will spit out lines as they are executed.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From eric at userland.com  Sun Aug 22 14:20:03 1999
From: eric at userland.com (Eric Kidd)
Date: Sun, 22 Aug 1999 14:20:03 -0400
Subject: More random python observations from a perl programmer
In-Reply-To: <000101beec3d$36204fc0$cb2d2399@tim>; from Tim Peters on Sat, Aug 21, 1999 at 09:25:25PM -0400
References: <37bec222@cs.colorado.edu> <000101beec3d$36204fc0$cb2d2399@tim>
Message-ID: <19990822142003.O693@moebius.dartmouth.edu>

On Sat, Aug 21, 1999 at 09:25:25PM -0400, Tim Peters wrote:
> > Copying things not producing a copy is well, not what non-serious
> > programmers immediately think of.
> 
> Patience, here.  People will repeat this until it sinks in <wink>:
> assignment never copies in Python.  It just binds a name to an object.  This
> is something many (by no means all) people do struggle with at first, but I
> think it's because it's *simpler* than what they're expecting, and it takes
> some time before they stop trying to apply the overly convoluted mental
> models they bring with them from other languages.  Persist, and the light
> will eventually dawn!

Well, I'm a LISP and Dylan programmer, so I expected things to work that
way. :-) But I've been teaching Python to a few non-programmers, and they
seem to figure out references pretty quickly. I have them type:

    x = [1, 2, 3]
    y = x
    print x
    y[0] = 10
    print x

If they look confused, I just say, "Well, you only created one list here.
The variables are just names for it." They say "Oh", grasp the concept, and
move on to the next topic.

References are very simple. It's unlearning what you already know about
variables that causes people trouble. Python itself is a very learnable
language, at least for complete novices.

> Python isn't particularly concerned about being obvious at first glance
> (*no* computer language could be deluded enough to claim that!); it's more
> concerned with notations that, once mastered, are very hard to forget or
> misuse by mistake.

Guido was very clever (at least I think so). I've taught a few people the
basics of programming, and Python makes things so easy for the student.
Python is so *tiny*, and people can learn it in chunks.

You can give somebody a listener, and tell her to type things. She can
study one thing at a time. It's as if each little bit of the language
(functions, variables, expressions, statements, data types) was designed to
be separable from every other bit.

Perl's not a bad teaching language, but novice programmers (at least in my
experience) can't seem to form a mental model of "how it all works." Larry
Wall, being a very smart linguist, would say that such a mental model is
unnecessary--you can speak perfectly comprehensible English without having
studied grammar. I think I disagree with Larry: programming's not quite
like a natural language, and some people benefit from a global
understanding of how things work.

Then again, it may have something to do with the people I've been
teaching. In almost every case, they want to learn programming (and/or
Linux) because computers were mysterious to them and they wanted to
discover some simple, general principles. Unix is good for this, and so is
Python. But back when I tried to teach people Perl, a funny thing happened:
they learned how to program (more or less), but they didn't ever jump up
and say, "Oh! Programming is simple!" Programming stayed mysterious for
them. Of course, this probably has more to do with my teaching than it does
with Larry Wall's fine work. :-)

Now, Tom undoubtably experiences Perl the way I experience LISP (but more
so, because he's a better hacker than I am): it's lucid and clear, and
every feature makes sense. There are no mysteries. But some of his Perl
snippets have made my skin crawl--I can understand what's going on, but
there are so many complicated concepts crammed into such a tiny space. It's
not something I could *explain* to a novice without taking a two-day
detour. C++ is the same way--you're trying to explain what a "parent class"
is, but then you need to explain "empty virtual destructors", which
probably requires a short detour to talk about vtables and how a C++
compiler works.

So, my new Python motto: "Teach people Python, and let sleeping dogs lie."
:-)

Cheers,
Eric




From info at pythonware.com  Tue Aug 17 09:47:35 1999
From: info at pythonware.com (PythonWare)
Date: Tue, 17 Aug 1999 13:47:35 GMT
Subject: ANN: PythonWare Sound Toolkit (PST) release 0.1
Message-ID: <007d01bee8b7$6ecbc5e0$f29b12c2@secret.pythonware.com>

The folks who gave you the Python Imaging Library (PIL) now
introduces yet another useful library:

The PythonWare Sound Toolkit (PST) reads sound files in a
variety of formats, and plays them on different hardware plat-
forms (currently Windows and Sun only).

The first, very rough, release is available from:

    http://www.pythonware.com/

PST 0.1 is shipped under a Python-style license, and can be used
without fee also in commercial projects.  For more info, see:

   http://www.pythonware.com/products/pst/

enjoy,
the PST development team




From fredrik at pythonware.com  Tue Aug 17 11:12:28 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 17 Aug 1999 15:12:28 GMT
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org> <14265.5096.537800.149227@buffalo.fnal.gov> <37B91E9A.2451936C@w3.org> <001301bee891$82fccc60$f29b12c2@secret.pythonware.com> <1Feu3.381$mV.141087@news.shore.net>
Message-ID: <000f01bee8c2$ec2b6940$f29b12c2@secret.pythonware.com>

Michael P. Reilly <arcege at shore.net> wrote:
> The better question is:
>    Why did it change from documenting these components in the OS-
>    dependant sections (specifically only posix) to documenting them
>    mostly in the module-os section in the current release?
> 
> Except for "New in version 1.5.2." comments, the pre-1.5.2
> documentation listed nothing more than exec[lv]* functions and there
> was only the "posix" module documented.

consider the original:

    it's called "unlink", and it's implemented by the "nt" module on
    your computer. look under "posix" for a description, and don't
    forget to access it via the "os" module to make sure your code
    is portable. (1.2)

which later become:

    while that function is actually implemented by the "nt" module
    on your computer, it's documented under "posix".  by the way,
    access it via the "os" module to make sure your code is portable.
    (1.5)

with:

    you'll find that function under "os" (1.5.2)

or Dan's preferred:

    you'll find that function under "os", unless it's a posix-
    specific function. (1.5.2+?)

</F>




From dworkin at ccs.neu.edu  Tue Aug 24 13:41:00 1999
From: dworkin at ccs.neu.edu (Justin Sheehy)
Date: 24 Aug 1999 13:41:00 -0400
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan> <slrn7s408d.gvn.wtanksle@dolphin.openprojects.net> <hhvha5tswc.fsf@forcix.roof.lan>
Message-ID: <vndaerht6tv.fsf@betelgeuse.ccs.neu.edu>

forcer <forcer at mindless.com> writes:

> You didn't quote the example where i pointed out the real
> problem with pass:
> 
> if foo:
>    # bar
>    # baz
> qux
> 
> E.g. if you're developing code and just commented out a (not yet
> worked out branch) to work on a different part of the program.
> Requireing pass here is unecessary.

For what you describe, pass is already unecessary.  You simply left
out one `#'.

Assuming that I have the following snippet:

if foo:
    bar
    baz
quux

If I want to do what I think you are describing above, I comment out
the entire if expression, not just its body:

# if foo:
#     bar
#     baz
quux

Is there any reason not to do it this way, or have I somehow missed
the point behind of your example?

I don't understand how `pass' has much to do with this, really.

-Justin

 



From hschroeder at gmx.net  Tue Aug  3 09:49:36 1999
From: hschroeder at gmx.net (Henning Schroeder)
Date: Tue, 3 Aug 1999 15:49:36 +0200
Subject: Events with pyKDE 0.8.
References: <7o6b0i$12ec@enews3.newsguy.com>
Message-ID: <g1s6o7.3v.ln@odin.henning-schroeder.de>

Peter Torstenson <p.t at iname.com> wrote:
> I have defined an event as follows.
> [..]
> Now. this works fine on the mainwidget. But how do I define it to work with
> a widget placed on the main widget. Ex, a label.
Look at the source below. You have got two possibilities to solve your problem.
Either you create you own label through subclassing QLabel and implementing
mousePresEvent or you can catch the events for a simple QLabel with 
installEventFilter. The second method is quite cool because you can use it with
any widget without sub-classing (I'm working on a dialog-editor using this
feature). Unfortunately you cannot (as far as I known) convert the QEvent to 
QMouseEvent in eventFilter. In C++ it is simple macro (definded in qevent.h)
but pyKDE does not support this (even not in pyKDE 0.9pre2 :-( ). This would 
be  necessary to access to mouse position.

Greets
  Henning


--------------------------------------------------------------
from kde import *

class MyLabel(QLabel):  
    def mousePressEvent(self, event):
	print "mouse button on %s pressed" % self

class Window(QDialog):
    def __init__(self, *args):
	apply(QDialog.__init__, (self,) + args)
	self.label1 = QLabel("Hello", self)
	self.label1.installEventFilter(self) # catch events
	self.label2 = MyLabel("World!", self)
	self.label2.move(self.label1.width(), 0)
	
    def eventFilter(self, object, event):
	if event.type() == Event_MouseButtonPress:
	    print "mouse button on %s pressed" % object
	    return 1 # event handled
	else: 
	    return 0 # pass event to next handler

import sys
app = QApplication(sys.argv)
win = Window()
app.setMainWidget(win)
app.mainWidget().show()
app.exec_loop()



From schorsch at schorsch.com  Sun Aug 15 06:55:50 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Sun, 15 Aug 1999 10:55:50 GMT
Subject: Regarding the magical "or" operator
References: <37b5d712@cs.colorado.edu>
Message-ID: <7p66bk$5ai$1@nnrp1.deja.com>

Tom Christiansen wrote:
[...]
> My question is this: have you ever had an urge for it (or a variant)
> to care about the difference between None and any other false value?
> I am certainly *not* asking for such a thing.  I'd just like to know
> whether anyone ever cared.
[...]

Yes, this has happened to me a few times. The usual setup is a
function that normally either returns a list or a numerical value,
but can fail in some circumstances, and where, obviously, [] or 0
are valid return values. In most cases, another look at the problem
will reveal a missed  opportunity to throw an exception for the
failed case.

There may be situations where an exception would at least *look*
like too much overhead. But with the current semantics, the extra
check for None is not really simpler or faster to type.
With your unproposed "nor" operator, the situation would change
a little. But then, even this would only apply to the special
case where the resulting value is used in an assignment.
As soon as other decisions depend on this result, the standard
exception handling seems to result in the most readable code,
especially in combination with the "else" and "finally" clauses
after "try" and "except".

And yes, I recently have started to use "or" and "and" as
assignment value selectors, as well as the abovementioned "else"
in exception handling and in looping constructs. It doesn't happen
very often, but can simplify the code a lot in some situations.
Given my lisp background, it's actually surprising that I lived
without noticing those possibilities for my first few years
of Python exposure...


Have fun!

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From wware-nospam at world.std.com  Fri Aug 13 15:42:25 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Fri, 13 Aug 1999 19:42:25 GMT
Subject: Python microthreads
Message-ID: <FGF5Ep.44z@world.std.com>

I have slightly improved my microthreads package.
It now has a web page:

    http://world.std.com/~wware/uthread.html

As before, the ftp download location is:

    ftp://ftp.std.com/pub/wware/uthread.tgz

The new version defines new microthread-safe versions of
a queue and a random number generator. It also removes
the hard-coded limit on the size of frame stacks, which
are now linked lists.

I have made very modest progress toward a merge with
Stackless Python. My microthread effort will become
obsolete in not too many months as Christian makes more
progress, so I'm planning to invest more of my time in
tinkering with applications.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From jstevens at basho.fc.hp.com  Fri Aug 13 23:04:21 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:04:21 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132309.RAA06899@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:09 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cea5@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :> Instead, you need to remember to use
> :> 
> :>     a.append("delta")
> :
> :According to what you said earlier, this is better than the Perl
> :way.  Appending to a list is better, 'cause you can tell the
> :difference between adding a new element, vs. reassignment: two
> :different operations, right?
> 
> Wrong.  I am merely sticking something at a particular place.
> If that place doesn't exist yet, MAKE IT SO.  Don't make 
> me check the length first.  

Your own reply indicates that these are two different operations.

In adding, you know you will not be overwriting and existing reference.

In assignment/reassignment, you don't.  When combining the two, you
are forced to add checks to determine the difference.

> :Excuse, but the two cases are different.  One is reassignment.  The
> :other is resize-list-and-add-a-new-element.  
> 
> Don't be ludicrous.

I'm not.

FYI: I have consistently treated you with respect.  I ask only
     that you do the same.  However, this reply makes me wonder.
	 You know darn well that adding something to a container
     object is is different than putting something into an
     existing slot.

> I don't care whether something is there before.

Which goes a long way towards explaining some of those Perl
defects I was refering to in other replies. . .

> Are you telling me that python should have one syntax for
> 
>     s = "original string"
> 
> and then a different one for
> 
>     s = "a new string that replaces the old one"

A variable is not a container class instance.  Your analogy is fatally
flawed.

> It doesn't do that for dictionaries.

An acknowledged flaw in Python's orginal dictionaries, one which has
been partially solved on extraction through the addition of the get
method.  Insertion remains to be fixed, as there is some philosophical
disagreement as to the correct semantic.

> It shouldn't do so for lists.

Lists, by definition, do not have "empty", or non-existent slots, so
overloading assignment could violate the semantic of "list".  Append
is both simpler, and less error prone.

So I must strongly disagree.  The existing system is more correct.

John S.




From garry at research.att.com  Thu Aug 26 15:18:04 1999
From: garry at research.att.com (Garrett G. Hodgson)
Date: Thu, 26 Aug 1999 19:18:04 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com> <xlx1zcqv4vj.fsf@gold.cis.ohio-state.edu>
Message-ID: <37C592EC.BFC9FC9@research.att.com>

Matt Curtin wrote:

> Tom complains loudly when something is broken or stupid.  Tom is very
> direct; some would say "abrasive".  Tom is almost always right.

debateable.

> The lack of Python man pages is a problem for Unix folks.

i've been a unix folk for decades, and it's not a problem for me.
i click in my browser, instead of typing "man".  big deal.

a simple shell (or perl) script can convert the html pages to
cat-able form.  i do this on occasion when no browser is handy.

> We have an
> online manual for a reason.  We're used to tools like `man' and
> `apropos' that use the manual and help us find the things we seek.
> Tom isn't "attacking" Python for not having it; he's (loudly) pointing
> out a very legitimate gripe.

yes and no.  it clearly bothers him, and perhaps others.  yet within
the python community there has been no such outcry.  this is similar
to the periodic "whitespace sucks" flamewars that erupt, typically
initiated by a non-user (sadly, most often a perl user).  for most
users, these are non-issues.  pointing out that "i wish there were
man pages; it would've made it easier if i could use familiar tools"
would've gotten a more useful response than tom got.

> Tom is trying to get the Perl community to learn Python and learn from
> it.  Python has done some things well.  It has done some other things
> not so well.  We, the users and designers of languages, have a great
> deal to learn from each other.

indeed.

> Perl folks generally seem to recognize
> that; it's sad that so many other language communities don't
> reciprocate.  It's especially sad to see that Python's is one that
> seems to like its glass house.

nonsense.  i've learned quite a bit here about other languages, in
many discussions.  the difference between those and the recent perl
thread is that the profitable ones tend to be stated as, "i find
feature X from language Y really useful.  does/could python
have this?"

rather than, "language Y is better than python.  it has feature X."

-- 
Garry Hodgson			comes a time
garry at sage.att.com		when the blind man
Software Innovation Services	takes your hand, says,
AT&T Labs			"don't you see?"



From fredrik at pythonware.com  Fri Aug 20 04:56:20 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 08:56:20 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu><19990819174800.D5227@xs4all.nl><37bc313b@cs.colorado.edu> <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us>
Message-ID: <019301beeaea$004c4d80$f29b12c2@secret.pythonware.com>

Jeremy Hylton <jeremy at cnri.reston.va.us> wrote:
> I did think about this before I jumped, but it still seems fair to
> jump.  As a possible counterexample to the "if it don't have manpages
> it don't exist" argument, I would suggest the C programming language.
> I don't believe there are manpages that describe the language itself,
> yet I found it (relatively) easy to learn.  The K&R C book is mighty
> fine. 
> 
> It's also worth noting that the old Lutz book is a little like the
> pre-ANSI K&R C book.  You're not going to find ANSI changes in the old 
> K&R, and you're not going to find Python 1.5 info in the old Lutz
> book.  In neither case does that mean the improvements don't exist.

as an aside, it's a bit sad that Tom didn't read the book's
preface, where Mark says, among other things:

    "... It's Not a Reference Manual ... This book is more like
    a tutorial than a reference manual. ... this book doesn't
    replace Python's reference manuals. If you want an ex-
    haustive list of language rules, the manuals do the job ...
    the manuals can be used to supplement the material
    here, if needed ..."

on the other hand, Webster says a "manual" is "a book that
is conveniently handled," so maybe Mark should have used
the word "handbook" to be on the safe side ;-)

</F> 




From fredp at multimania.com.nospam  Tue Aug 31 12:06:57 1999
From: fredp at multimania.com.nospam (Fred Pacquier)
Date: 31 Aug 1999 16:06:57 GMT
Subject: Newbie (read: pot. stupid) question on objects/namespaces
References: <8E339DC2FPaCmAnRDLM@news.dial.oleane.com> <m3iu5w5an5.fsf@pcrm.win.tue.nl> <8E33ADDF0PaCmAnRDLM@news.dial.oleane.com> <m3emgk55ic.fsf@pcrm.win.tue.nl>
Message-ID: <8E33B4AE7PaCmAnRDLM@news.dial.oleane.com>

stephan at pcrm.win.tue.nl (Stephan Houben) said :

>> Duh. Is this a total "no-no", or what ? If this is only for a handful
>> of instances that are supposed to persist through the entire program
>> lifetime, is it okay to explicitely del() them at the end or not ?
>
>No, this was just a word of caution!
>If it's just a handful of instances, don't bother.

Exactly 18 :-)

>However, if you want to get rid of them, del'ling them doesn't work.
>You must break the circularity to get them garbage collected.
>In my original example:
>> >class Baz:
>> >    def __init__(self):
>> >        self.x = MyClass(self)
>,when you have created an instance:
>baz = Baz()
>,you can break the circularity by either doing:
>baz.x = None
>or
>baz.x.owner = None
>But don't do this in the __del__() method of Baz!
>That's not good enough. You have to do this in order to
>get __del__() called in the first place.

Got that. Once again, thanks !

>Now, if you use JPython, you have real garbage collection, and
>all this is a non-issue.

Er, no. One thing at a time :-)

-- 
YAFAP : http://www.multimania.com/fredp/



From zajac at nortelnetworks.com  Thu Aug 19 10:43:13 1999
From: zajac at nortelnetworks.com (Ron A. Zajac)
Date: Thu, 19 Aug 1999 09:43:13 -0500
Subject: NEWBIE: Setting WinNT python to point to my tcl DLL
Message-ID: <37BC1801.3404C9E8@nortelnetworks.com>

Dear Python Anciens,

  I've installed the latest Python on my WinNT system, but declined the
Tk install, since I already have tcl/Tk8.1.1 on my system.  It appears
to be the case I need to point the system to my current tcl/Tk install,
however.  "IDLE" can't find "tcl80.dll" which, of course, doesn't exist.

  Where can I make this change so IDLE can find the relevant DLL?






From fdrake at acm.org  Thu Aug 19 15:24:15 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 19 Aug 1999 19:24:15 GMT
Subject: More random python observations from a perl programmer 
In-Reply-To: <14268.21198.934117.241668@anthem.cnri.reston.va.us>
References: <37BC6986.A66B044E@starmedia.net>
	<199908191634.KAA01333@jhereg.perl.com>
	<14268.21198.934117.241668@anthem.cnri.reston.va.us>
Message-ID: <14268.23007.239507.166209@weyr.cnri.reston.va.us>

Barry A. Warsaw writes:
 > give poor overworked Fred the most bang for the buck.  Info format
 > used to be supported, but I don't think that was used by enough people

  Bucks?  I'm supposed to be getting bucks for this?
  Check the FTP site, Barry; GNU info is there thanks to the active
Python community.

 > to justify the work involved in keeping it going.  Manpage format just
 > doesn't seem to be that popular either.  I know that sucks for you

  Manpages have never been requested, unless you consider Tom's
complaint a request.  I've received nothing at
python-docs at python.org.  Certainly not a contributed conversion
script.

 > Tom, but I'm sure Fred would gladly take a contribution that generated
 > manpage format from the latex source.  You might even be surprised at
 > the languages he'd accept such a contribution in! :)

  Hey, you're making me look bad!  ;-)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From fredrik at pythonware.com  Sat Aug 14 14:06:35 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sat, 14 Aug 1999 18:06:35 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <934580930snz@vision25.demon.co.uk>
Message-ID: <000b01bee67f$bfdd2990$f29b12c2@secret.pythonware.com>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
> Cos Guido doesn't like them, I imagine.

seen on the python list, back in november 1991:

    someone wrote:
    > How about providing assignment operators?

    guido:
    > Agreed.  Now that the parser can actually recognize
    > two-character operators, adding these wouldn't be so
    > hard (except that the grammar tables will grow again :-( ).

since this hasn't made it into the language yet, I suppose
keeping the grammar tables small is quite important ;-)

</F>




From amaranda at spider.com  Mon Aug 16 05:33:36 1999
From: amaranda at spider.com (Alex Maranda)
Date: Mon, 16 Aug 1999 10:33:36 +0100
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au>  <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <37B7DAF0.3198@spider.com>

Sam Holden wrote:
> I know a fair few pythoners in fact, I even like some of them. What I meant
> was that anyone who would do such a thing with perl deserves to be killed.
That's the only bit which bothers me; would you put your oppinions in
practice given the chance? (morality and law put aside of course :-)

> Also don't tell anyone but I actually like python a lot. I don't use it much
> since I have more experience with perl, and more importantly I know the
> modules available for perl better, and can quickly search for one to
Get a copy (it's worth printing out, believe me) of the 'Python Library
Reference'.
> do whatever I need next.
That's good to hear. Perhaps you're in the position (or will be in time)
to make a fair comparison between Perl and Python, as coming from an
experienced and unbiased Perl programmer. The same comparison done from
the other side of the mirror (i.e. Python) would be great also. And best
of all, surveys ref what are the language(s) used for and not only how
many people are in each camp...

> >> use impythonate; # STILL NEED THAT ONE SEMICOLON, DAMMIT!
> >>
> >> for $i (1..10)  # COMMENTS ARE OKAY
> >>   print "$i: "
> >>   my $isq =  \
> >>   $i **2 # LINE CONTINUATIONS WORK AS IN PYTHON
> >>   print " $isq\n"
> >>
> >> print "done\n"
> >I know Python fairly well (but I'm no expert) and I can easily read this
> >piece of Perl.
> 
> But it would be no harder to read with ; on the end of each statement and
> a pair of {}s.
Fair enough; the loop is so simple I would understand it in COBOL also
(no, I don't know COBOL :-)

> 
> Working out that strings interpolate should be a bigger leap than working
> out that blocks have {}s around them.
If this all impythonate is doing it's not worth the trouble of course.

> Everyone seems to misunderstand the whole concept of TMTOWTDI. It doesn't
> mean all ways are equal. That is a way of writing perl which works, it is
> however not very maintanable. The only way to determine what syntax is
> valid is to know perl's syntax, and to be able to read and understand the
> impythonate.pm code - since there is no documentation and it is _not_ the
> same as python.
> 
> Writing it in 'normal' perl reduces the requirements to understanding, and
> speeds it up to boot.
Fair enough again; I misread your post (or I read to much into it), and
I thought you're against Python's syntax and whitespace in general. If
it doesn't make sense for Perl, that's fine with me.

> >A couple of days ago I read a bit in Bugzilla's README which amused me:
> >"A computer which doesn't have Perl installed is a very sad computer
> >indeed". Your post is frightening (unless it was a Perl-style joke).
> 
> How was it frightening?
I don't follow comp.lang.perl.* so I don't know what's accepted behavior
there. Imagine my shock when encountering the word 'kill' as punishment
for programming in Python style, when reading my favorite newsgroup.

Regards (no doubts this time),
Alex



From tchrist at mox.perl.com  Sat Aug 14 14:35:28 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 14 Aug 1999 11:35:28 -0700
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov>
Message-ID: <37b5a8e0@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, Drew McDowell <drew.mcdowell at msfc.nasa.gov> writes:
:I looked in the FAQ and couldn't find anything on the Compound Assignment
:Operators.  (+=, *=, -=, etc..)  Why dosen't Python support them?
:Is there a easy way to add them?

Well, to start with, Python does not even *have* an assignment operator.
Nope, not one.  Assignment is a statement, not an operator.  That's while
you can't write:

    while line = file.readline():
	print line

for example.

Mostly it's the old "burnt child fears fire" syndrome.  Or, if you would,
"fool me once, shame on you; fool me twice, shame on me".  By outlawing
it, you can't get burned/fooled by it.  Too many horrendous C programming
errors happen from confusing:

    while (var = function()) { }

with 

    while (var == function()) { }

Think of the classic loop:

    while ((ch = getchar()) != EOF) { } 

You end up having to use parentheses in do group the assignment,
and then test its result.

As for why you can't write the statement (note: not the expression)

    x["every"]["good"]["boy"]["does"]["fine"] =  (
	x["every"]["good"]["boy"]["does"]["fine"] + 1 )

as 

    x["every"]["good"]["boy"]["does"]["fine"] += 1

The basic reason seems to be that there are some people who believe that
this kind of thing is cruel and unusual punishment levied upon the poor
Pascal(ish) programmers purely for sake of brevity.

Common arguments in support of such ops include:

    1) This requires care to avoid mis-duplication:
	x["every"]["good"]["boy"]["does"]["fine"] =  (
	    x["every"]["good"]["boy"]["deos"]["fine"] + 1 )
    2) Side effects must be duplicated:
	x[every()][good()][boy()][does()][fine()] =  (
	    x[every()][good()][boy()][does()][fine()] + 1 )
    3) It's too long to type.
    4) Needless redundancy obscures the real meaning.

Whereas common rebuttals include:

    1) Toss /bin/ed and get yourself a real editor.
    2) Don't use side-effects, you wicked creature, you!
    3) It's more efficient to keep a tmpobj anyway.
    4) Redundancy can clarify through redundancy.

As to why there's no x++ there, consider how often people from a
Pascal(ish) background are confused between x+1 and x++ and ++x, and
you'll begin to see what kind of problems caused it to be vetoed.

But the best reason is probably the one another poster already supplied:
because Guido didn't like them.  An interesting study would be to check
the extent to which he also forbids such constructs in his own C code. :-)

--tom
-- 
    Besides, REAL computers have a rename() system call.    :-)
                    --Larry Wall in <7937 at jpl-devvax.JPL.NASA.GOV>



From greg.ewing at compaq.com  Tue Aug 31 22:52:12 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 01 Sep 1999 14:52:12 +1200
Subject: P?G (Re: A Plea For A Return To Sanity!!!)
References: <000301bef12c$1bfd6940$ad2d153f@tim> <7q985l$7bi$3@newshost.accu.uu.nl>
Message-ID: <37CC94DC.D6DCA4B4@compaq.com>

Martijn Faassen wrote:
> Tim Peters <tim_one at email.msn.com> wrote:
> > [Bit Nik]
> >> Another quick question, are there such things as PUGs?
> > SIGs -- Special Interest Groups.
> No, no, SIGs something else again. He was asking about PIGs.

While we're at it, let's define the acronyms PAG (Python Appreciation
Group), PEG (Python Evangelism Group) and POG (Python... ummm...
any ideas?)

Greg



From philh at vision25.demon.co.uk  Tue Aug 10 19:00:02 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 10 Aug 99 23:00:02 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B02965.B1FE0E8D@prescod.net>
Message-ID: <934326002snz@vision25.demon.co.uk>

In article <37B02965.B1FE0E8D at prescod.net>
           paul at prescod.net "Paul Prescod" writes:
> Have you looked into XUL, 

No ...

> XwingML

... no ...

> and Glade?

... and yes.
 
> http://www.oasis-open.org/cover/xul.html
> http://www.bluestone.com/xml/XwingML/
> http://glade.pn.org/
> 
> I don't know whether these are exactly the same as what you are doing
> but they are certainly related -- and both already have implementations.
> Glade's is already in Python!

Glade is a GUI application that allows you to build a window visually.

My system (provisional name: ``Parrot'') is a command-line application
that takes an ascii file as input, and outputs a file containing source
code in your favourite language, to build the GUI.

> Phil Hunt wrote:
> > 
> > It is important to me that the tool is easy to use. This is because
> > I intend to use it myself!
> 
> Obviously ideas of ease of use vary. Some might say that XML is easier
> to use because:
> 
>  * all major text editors have a mode for it

all major text editors handle ascii well, too!
 
>  * it can be trivially parsed with existing code written in several
> languages

True; I don't think a parser for Python will be too difficult to 
write: my provisional spec only has 12 productions.

>  * APIs for tree manipulation are well known

True, but writing classes implementing recursive tree-structures
is almost trivial in Python. (One reason I like it).

>  * the syntax is (becoming) familiar

This is a good point.

>  * XML transformation languages (including some nice Python libraries)
> can convert your language into XUL and vice versa -- or whatever else
> comes along (and I have a feeling several of these interface description
> languages will come along ... there will likely be as many as there are
> widget toolkits)

What's XUL?

> You are certainly free to do it however you want. It's your language and
> it should be something you are comfortable with!

I will stick with my current format; but adding a feature to 
import/export to XML shouldn't be too difficult -- then I'd have
the best of both worlds.

> > It will also use the C preprocessor, so for example one could
> > write a macro for a label and associated textfield:
> > 
> > #define labelTF(txt,id,size) \
> >    label @id##Label txt textField @id cols=size
> 
> XML has a similar facility but I think that textual preprocessing is a
> little gross whether done in C or XML. I would encourage you to consider
> structured alternatives. Whether you use C-syntax or XML syntax, you
> could rather have a first-class "object definition" syntax. SVG has such
> a thing for defining 2-dimensional shapes.

I am intending to do something like this. For example:

   sub.font = "Helvetica"

would mean: all components under the current component default to
using the Helvetica font. Similarly:

   all.button.bcol = "111133"

would make the background colour of all buttons default to dark blue.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From achrist at easystreet.com  Thu Aug 26 22:17:33 1999
From: achrist at easystreet.com (Al Christians)
Date: Thu, 26 Aug 1999 19:17:33 -0700
Subject: Fuzzy string matching?
References: <199908261746.TAA25440@dionysus.fw.cuci.nl> <WXgx3.1646$v6.107045@typhoon.nyroc.rr.com> <y0j4shmux5a.fsf@vier.idi.ntnu.no>
Message-ID: <37C5F53D.B201DD1B@easystreet.com>

I've gotten good results with ad hoc algorithms using a longest common
contiguous substring routine.  There is an algorithm in the _Algorithms_
book by Rivest, et al, that produces the longest common non-contiguous
substring, which might be a better indicator of a match, but modifying
it to test for only contiguous substrings improves efficiency much,
particularly space efficiency.  The lengths of the 
two or three longest common contiguous substrings give a pretty good
indication of the degree of match in the applications I've tried (name
and address cleanup).   How to combine these lengths into a scalar
measure of match is the really ad-hoc part of it.  


Al



From vijaym at netscape.com  Wed Aug 18 00:35:31 1999
From: vijaym at netscape.com (vijay Mallela)
Date: Tue, 17 Aug 1999 21:35:31 -0700
Subject: Python and Oracle
Message-ID: <37BA3813.2BEB45A8@netscape.com>

Is there any current work going on with the Oracle module ?
Thanks
v




From tbryan at arlut.utexas.edu  Fri Aug 27 12:15:03 1999
From: tbryan at arlut.utexas.edu (Thomas Bryan)
Date: Fri, 27 Aug 1999 11:15:03 -0500
Subject: A Plea For A Return To Sanity!!!
References: <000001bef041$84deffa0$592d153f@tim> <37C68564.15207535@callware.com>
Message-ID: <37C6B987.1662D46F@arlut.utexas.edu>

Ivan Van Laningham wrote:
> 
> Tim Peters wrote:
> >
> > Besides,
> > Tom's very public misunderstandings are a *great* base on which to build a
> > "Python for Perlers" FAQ.  Python isn't self-evident at first glance, and > > it's good to relive the initial confusions through his articulate eyes 
> > <ahem> (you may not have liked how he phrased things, but there was no
> > doubt about what he was saying -- that's good writing).
> 
> Quite seriously, I think this is a very good idea.  If I knew more than
> nothing about Perl, I would volunteer to put such FAQ together; but not
> only do I know zilch, I don't even have time to be writing this
> letter ;-)

Is anyone putting this together?  Should it be a FAQ Wizard or what?
I could set up a FAQ Wizard on my Starship page if others are willing 
to contribute.  I must confess that I haven't used Perl much since
learning
Python, so I'd need a lot of help.

-------------------------------------------
Tom Bryan
Applied Research Laboratories
University of Texas at Austin



From jblaine at shell2.shore.net  Thu Aug 19 13:36:31 1999
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Thu, 19 Aug 1999 17:36:31 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7ro9ns.h6v.jblaine@shell2.shore.net> <37bc2ee5@cs.colorado.edu>
Message-ID: <slrn7rog4u.hri.jblaine@shell2.shore.net>

>In comp.lang.python, 
>    jblaine at shell2.shore.net (Jeff Blaine) writes:
>:If you put me on any UNIX machine without man pages containing
>:descriptions and command-line usage of the tools there, I would
>:be royally pissed.
>
>Exactly.  And I thought my statements were rather subdued for
>being royally pissed. 

Heh.  Alright...it's just that you and *at least* myself are coming from
very different views and usage expectations on this.  Perl didn't always
have pod, and more to the point Perl did not have pod when I learned it.
Perl had a gigantic man page that was so cumbersome to use, I didn't.

I always used the HTML version that used to exist at ufl.edu and it didn't
bother me in the least.  As a matter of fact, I found it much more useful
than any existing Perl documentation.

Because of that, I've never expected language reference to be available
in man page form on any UNIX box, with the exception of C function
reference which has just been there from day one for me.  So, I honestly
never blinked an eye when Python's docs were in HTML format, especially
given its wide platform availability (specifically non-UNIX) from
early versions.  I installed the documentation tarball, pointed lynx and
GUI browsers at it and that was that.

Even *when* Perl got pod, I didn't use the man pages due to their flat
and linear nature.

I know what version of Python I am running, and I know where the docs
for that version are.  Using Python 1.5.x and expecting a book which
is several years old (and based on 1.4) to be your source of
day-to-day reference seems pretty silly if you ask me.

As a royally-pissed-if-expected-tool-man-pages-are-not-there kinda
person, it doesn't bother me at all if a 3rd party language gets installed
and does not have man pages for each function or functionality.  I
can certainly agree that SOME sort of man page for 'python.1' would
make some sense, and more than that would only make more people happy...

Now, having said ALL of that, recently Transarc/IBM decided to STOP
doing man pages for AFS.  Man pages that I knew existed and depended
on.  This removal of service is the kind of thing that can really piss
a person off, and that seems to be what you're feeling coming from your
expectations.  Tom Christiansen believes man pages should exist for
any runnable command on a UNIX box (and ideally more than just that)
(right?).  Tom Christiansen does not find any for Python.  Tom
Christiansen is pissed.

Anyway, I seem to be entirely unable to quickly and concisely make a
statement anymore :)  I guess I more or less agree with you, I just
don't feel as strongly about it as you do for many reasons which are
spelled out in both of our messages.

>:I don't know, your little jab there in an otherwise great reference
>:document posting seemed completely petty to me.  It makes me think
>:you stuck it in there as a little "See, I'm dissing them like I am
>:supposed to!" due to the fact that you posted the message to both
>:c.l.perl and c.l.python.
>
>Don't be silly.  That wasn't that at all.

OK, sorry.  I was really hoping that wasn't the case.

>That's why we have "pod" as base docs, which can convert into HTML,
>text, info, latex, frame, troff (manpages), etc.  It makes it easy to
>run grep or find programs on.  This is very important.  And each module
>automatically installs that module's manpage (and optionally html doc,
>but this is less popular amongst perl folks).  That gets it added to
>the apropos and whatis and man commands automatically.  This is about
>0.002% of what the MakeMaker thingie in Perl does, but it's the part
>that matters for docs.  I've surely heard Python folks drooling over it
>for a long time.

Indeed, something like pod would be great, and many Pythoners definitely
do agree.

And before someone suggests gendoc, I've tried to get 'gendoc' and whatever
it used to be called to simply install and function properly 3 separate
times now over the last 2 years and have given up after an hour of work
and editing source code each time.



From gregm at iname.com  Wed Aug 25 00:42:03 1999
From: gregm at iname.com (Greg McFarlane)
Date: Wed, 25 Aug 1999 14:42:03 +1000
Subject: Tk's default Toplevel - counterintuitive
In-Reply-To: <5lso5ese9x.fsf@eric.cnri.reston.va.us>; from Guido van Rossum on 20 Aug 1999 at 10:44:10AM
References: <37BC6F2D.30A7@cs.bham.ac.uk> <37BC8D05.142E1DB0@aw.sgi.com> <37BC9CE8.67097FC6@vision.arc.nasa.gov> <37BCD6CF.B7A186BC@montana.com> <5lzozmskm3.fsf@eric.cnri.reston.va.us> <00dd01beeb11$e41a68e0$f29b12c2@secret.pythonware.com> <5lso5ese9x.fsf
Message-ID: <19990825144203.50611@nms.otc.telstra.com.au>

On 20 Aug, Guido van Rossum wrote:
> "Fredrik Lundh" <fredrik at pythonware.com> writes:
> 
> > Guido van Rossum <guido at cnri.reston.va.us> wrote:
> > > root.protocol("WM_DELETE_WINDOW", root.destroy)
> > 
> > how about making the destroy variant the
> > default?
> 
> I can't see any reason not to do it; I believe this is the default in
> Tk anyway, but Tkinter's destroy() does a little more than Tk's
> internal destroy.  Unless I hear a clamor of complaints I'll go ahead
> and do this.

As some proof that making it the default is "a good thing", Pmw has
been doing this, behind the scenes, for some years.  No one has
complained (or clamored) yet :)

-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
                   of precipitation in some areas.



From kistler at fnmail.com  Sat Aug 21 22:14:50 1999
From: kistler at fnmail.com (Per Kistler)
Date: Sun, 22 Aug 1999 04:14:50 +0200
Subject: Date Calculations
Message-ID: <37BF5D1A.9865CDF@fnmail.com>

Hi All

I'm making some Date Calculaton class with things like
deltaDays, addDays, daysInYear and so on, above the calendar 
and time classes. After some 300 lines I thought, I better
ask, whether that has already been done by someone else.
How is this?

Regards, Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From jpl at digex.net.nospamplease  Thu Aug 19 17:47:34 1999
From: jpl at digex.net.nospamplease (John Landahl)
Date: Thu, 19 Aug 1999 14:47:34 -0700
Subject: No man pages? (was Re: More random python observations from a perl programmer)
In-Reply-To: <199908192043.OAA07181@jhereg.perl.com>; from Tom Christiansen on Thu, Aug 19, 1999 at 02:43:26PM -0600
References: <19990819123502.A28466@digex.net> <199908192043.OAA07181@jhereg.perl.com>
Message-ID: <19990819144734.C28466@digex.net>

On Thu, Aug 19, 1999 at 02:43:26PM -0600, Tom Christiansen wrote:
> >What's all this about no Python manpages?
> 
> $ man python
> No manual entry for python
> [Exit 1]
> 
> $ locate python.1
> [Exit 1]
> 
> That there appears to be running a 1.5.1 version on a Redhat
> Linux system.  I haven't tried a from-scratch build.  I'm
> happy to blame RH.  Their doc situation is nasty.

Ahhhhhhhhhh, that seems to be the problem.  I just tried it on a recently
built (RedHat)SparcLinux box and got the same thing.

Having followed the manpage discussion a bit, I see two arguments that might
have gotten confused by this new info:

  1) There should at least be a python(1) manpage that tells you command
     line args, some basic information about the interpreter, etc.

  2) There should be manpages documenting all aspects of the language.

If your problem was mostly with 1) then I concur, and so would most Unix
types, I think.  The rest of this debate seems centered around issue 2), and
I'm not sure that was your main point.  I agree that 1) should be the
minimum, but I would tend to look for web-based cross-referenced
hypertextified documentation for more detailed information.  Though I've
definitely found "man perlfunc" to be extremely useful when I want to get at
something quick, so I agree that Python manpages would be a nice addition.

- John Landahl / jpl at digex.net



From I.Clarke at NOSPAM.strs.co.uk  Fri Aug  6 11:57:52 1999
From: I.Clarke at NOSPAM.strs.co.uk (Ian Clarke)
Date: Fri, 06 Aug 1999 16:57:52 +0100
Subject: Python GUI
References: <19990805141904.115.qmail@hotmail.com>
Message-ID: <37AB0600.869DFD87@NOSPAM.strs.co.uk>

> > > Is there a Python GUI builder out there anywhere?
> >Not that I'm aware of.
> >Anyone know of one in-the-works?
> http://www.pythonware.com/products/works/

Unfortunately it doesn't seem to be open source :-(

I have often thought that a Python based IDE would be an ideal
OpenSource answer to Visual Basic.

Ian.

-- 
Remove the "NOSPAM." from my email address before replying.



From tim_one at email.msn.com  Sat Aug 21 04:09:15 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 21 Aug 1999 04:09:15 -0400
Subject: UnboundLocalError (was RE: Compound Assignment Operators ( +=, *=, etc...))
In-Reply-To: <14269.28496.156064.652501@weyr.cnri.reston.va.us>
Message-ID: <000101beebac$75da5de0$0c2d2399@tim>

[Fred L. Drake, Jr.]
>   Current behavior (1.5.2 as released) raises UnboundLocalError (a
> subclass of NameError), which is much more informative and is
> usually exactly what is needed to point out to the programmer
> where the coding error is.

UnboundLocalError is in the CVS development tree, but is not in 1.5.2.

Tom noted the dangers of separating the doc release from the code release,
so he'll be pleased to see the point has been illustrated <wink>.  I'll add
that there's also a downside to the public CVS tree:  many of us corporate
apologists don't run the released 1.5.2 anymore, so say stuff that isn't yet
true for those who are.  Perl's corporate apologists have been doing this
for years, though <wink>.

the-world-is-fraught-with-teensy-perils-ly y'rs  - tim





From stephan at pcrm.win.tue.nl  Tue Aug 24 03:18:17 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 24 Aug 1999 09:18:17 +0200
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> 	<m3vham44g5.fsf@atrus.jesus.cam.ac.uk> 	<14257.54137.72005.705719@anthem.cnri.reston.va.us> <37b6f1c2.347828751@news.triode.net.au>  <37B82032.E7A11398@starmedia.net> <slrn7rpbk8.7pl.neelk@brick.cswv.com> <37C1E3D6.CD835CD1@compaq.com>
Message-ID: <m3g1194pfq.fsf@pcrm.win.tue.nl>

Greg Ewing <greg.ewing at compaq.com> writes:

> Andrew Csillag <andrew at starmedia.net> wrote:
> >
> >Also, you may want to look at how the SmallEiffel compiler does it's
> >type inferencing...  
> >(and obviously the souce code is available to give you an implementation
> >>to look at -- written in Eiffel).
> 
> Are you sure about the source code? Last time I looked
> at SmallEiffel, they didn't give you the Eiffel source,
> only the generated C code.
> 
> Has this changed? If so, it's very interesting.

It has. Full Eiffel source is now available.

SmallEiffel is now the official GNU Eiffel compiler.
So obviously source has to be available.

Greetings,

Stephan



From yishai at platonix.com  Tue Aug 17 14:10:15 1999
From: yishai at platonix.com (Yishai Beeri)
Date: Tue, 17 Aug 1999 21:10:15 +0300
Subject: Help with Threads and httplib please
Message-ID: <37B9A587.3160428E@platonix.com>

Hello,

I have a weird problem running several python threads that use httplib
to get some pages from the web. all runs ok, until suddenly all the
threads get stuck. A ctrl-c makes them all continue, until the next time
they're stuck again.

>From what I saw in the debugger, it seems that one thread is stuck in
the "connect" of the socket under httplib, and all the other threads are
waiting for some reason.

Has anyone seen this problem or has an idea? I am using Linux (RedHat 6)
and Python 1.52 with gcc 2.95

Thanks,
Yishai Beeri





From bowman at montana.com  Fri Aug 27 23:36:59 1999
From: bowman at montana.com (bowman)
Date: Fri, 27 Aug 1999 21:36:59 -0600
Subject: A scrollable Frame or Canvas widget
References: <37C73097.8165F42C@uniserve.com>
Message-ID: <37C7595A.351052FC@montana.com>

Justin Muir wrote:
> 
> Could anyone out there also clear up the mechanics of the different
> callbacks/commands invoked by the arrows and the scrollbar?

Quick and dirty Text widget with two attached scrollbars. There probably
are cleaner and better ways to do the job, but this may help.

from Tkinter import *

def main():
	root = Tk()

	f = Frame(root)
	text = Text(f, width=40, height=10, setgrid='True', wrap='none')

	xscroll = Scrollbar(f, orient=HORIZONTAL, command=text.xview)
	yscroll = Scrollbar(f, orient=VERTICAL, command=text.yview)
	xscroll.pack(side=BOTTOM, fill=X)
	yscroll.pack(side=RIGHT, fill=Y)
	text['xscrollcommand'] = xscroll.set
	text['yscrollcommand'] = yscroll.set

	text.pack(side=LEFT, fill=BOTH, expand='True')
	f.pack(side=TOP, fill=BOTH, expand='True')

	root.mainloop()

if __name__ == '__main__' :
	main()


-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From rjroyNoSpam at magma.ca  Thu Aug 12 09:43:41 1999
From: rjroyNoSpam at magma.ca (Robert J. Roy)
Date: Thu, 12 Aug 1999 13:43:41 GMT
Subject: Problem when using FileDialog from PMW Dialog.  Quick fixes?
References: <37B21CE1.488F45FA@vision.arc.nasa.gov>
Message-ID: <37b2cb01.501756387@news1.on.sympatico.ca>

What you are trying to do is common practice.

If this code snippet is accurate, the problem is in your use of global
dialog. In CreatePMWDialog you assign dialog to the Pmw.Dialog
then in ListFiles, you chage the binding of the global dialog variable
to point to the FileDialog. So your subsequent call to DialogAction is
trying to act on a non-existing widget. 

If you changeListFiles to the following, everything works fine. 

def ListFiles():
    global dialog
    dlg = LoadFileDialog(dialog.parent)
    file = dlg.go("")
    print file

You might want to look at using tkFileDialog instead. It brings up the
standard file dialogs, presenting a more familiar interface to Windows
users.

check out http>//www.pythonware.com/library/tkinter/introduction/ for
more info on how to use. (Sorry I can't give you more info, I can't
get through to pythonware just now)

Good Luck!
Robert Roy




On Thu, 12 Aug 1999 01:01:21 +0000, Chad Netzer
<chad at vision.arc.nasa.gov> wrote:

>I have an application where I open a Python Mega Widget dialog (local,
>not global) and within it I later open a FileDialog() for file selection
>(Don't ask, why, but calling a nested dialog makes sense for the app.)
>In any case, it doesn't work.  When the FileDialog exits, the Pmw Dialog
>cannot get focus back (even with some code to try to force it.)
>
>Before I pursue this further, can anyone tell me if it is even worth
>trying?  It is nice to use the canned FileDialog, but its implementation
>is somewhat crufty (IMHO) and I can probably work around it in other
>ways (Creating my own FileDialog from PMW widgets is a nice possibility.
>)   But if there is a quick fix, I'd like to hear it.
>
>Below you'll find  some psuedo-code to show roughly what I'm doing.  Run
>the code, click on the "List" button to activate the FileDialog, then
>exit the FileDialog.  Hitting the PMW Dialog "Cancel" button doesn't
>work (as it does if you never open the FileDialog).  Pardon the rough
>code :)  You will probably have to kill the process if you trigger
>the problem, so be warned.
>
>Chad Netzer
>chad at vision.arc.nasa.gov
>
>------------------------------ [ Cut Here ]
>------------------------------
>from Tkinter import *
>import Pmw
>from FileDialog import LoadFileDialog
>
>def CreatePMWDialog(parent):
>    global dialog
>    dialog = Pmw.Dialog(parent,
>                        buttons = ('OK', 'Cancel'),
>                        defaultbutton = 'OK',
>                        title = 'A test: The FileDialog will steal focus
>forever',
>                        command = DialogAction)
>    dialog.parent = parent
>
>    interior = dialog.interior()
>    entry_frame = Frame(interior)
>    list_but = Button(entry_frame, text="List", command=ListFiles)
>
>    list_but.grid(row=0, column=1, sticky=N+S)
>    entry_frame.pack(side=TOP, anchor=NW)
>
>    # Now execute the dialog
>    dialog.activate()
>    return
>
>def ListFiles():
>    global dialog
>    dialog = LoadFileDialog(dialog.parent)
>    file = dialog.go("")
>    print file
>
>def DialogAction(button_name):
>    global dialog
>    if button_name == 'Cancel':
>        dialog.deactivate()
>    elif button_name == 'Ok':
>        dialog.deactivate()
>    return
>
># Test it by parsing a prm file
>if __name__ == '__main__':
>    tk_root = Tk()
>    foo = CreatePMWDialog(tk_root)
>    tk_root.mainloop()
>
>




From philh at vision25.demon.co.uk  Mon Aug 16 18:21:08 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 16 Aug 99 22:21:08 GMT
Subject: Anyone else find Tk akward?
References: <37B34DBB.E5A79760@sprint.ca> <NeMs3.6611$gO1.175350@news2.giganews.com> <37B87BAC.CE092017@compaq.com>
Message-ID: <934842068snz@vision25.demon.co.uk>

In article <37B87BAC.CE092017 at compaq.com>
           greg.ewing at compaq.com "Greg Ewing" writes:
> Christopher Browne wrote:
> > 
> > The Tk model of packing "objects" onto a "canvas" allows it to
> > self-optimize, and lets the programmer not need to worry so much about
> > those sorts of fiddly details.
> 
> This is good, except that if you just pack things
> together in the simplest and most obvious way, the
> results are awful -- everything is jammed together,
> and nothing resizes the way you want. To get a layout
> which looks good and works properly, you need to deal
> with another set of fiddly details, namely the various
> padding, filling and expanding options, which (for
> me, anyway) seem to work in very unintuitive ways.
> 
> The idea of an automatic layout manager is as good
> one, but I don't think that any of Tk's geometry
> managers are a good implementation of it. I'm not
> surprised that some people find them frustrating.

I'm currently in the middle of (or to be more accurate, near the
start of), writing a system that will hopefully allieviate this
probelm -- it's a GUI design language with multiple back ends,
one of which will be Python/tkinter.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From nospam at bitbucket.com  Wed Aug 25 03:57:47 1999
From: nospam at bitbucket.com (Phil Mayes)
Date: Wed, 25 Aug 1999 00:57:47 -0700
Subject: UML
References: <37B96A3A.796BB034@compuserve.com>
Message-ID: <mmNw3.12$z3.2904@newsfeed.avtel.net>

robert_wilhelm_land wrote in message <37B96A3A.796BB034 at compuserve.com>...
>Are any persons in this group using the unified modelling language?
>If yes, could you kindly explain when to use the UML and what the
>advantages are?
>I have read some articles, but I'm not able to analyse the sense of the
>UML.
>
>Would be grateful for some help

UML is a standardised set of graphical representations promulgated by the
gurus who specialise in selling high-priced snake-oil solutions to big
organisations.  They have simply brought together a number of traditional
methods such as state diagrams and decreed the detailed format.  To judge
their skill level, consider this: they chose to draw their class
inheritance trees with the arrows going up towards the root.

It may be a useful job skill when interviewing at a large corporation,
but IMHO, a whiteboard and 60 minutes thinking wins hands down.
--
Phil Mayes    pmayes AT olivebr DOT com







From tim_one at email.msn.com  Tue Aug 31 22:10:35 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 31 Aug 1999 22:10:35 -0400
Subject: Soundex implementation (was: RE: Fuzzy string matching?)
In-Reply-To: <000801bef419$3f5be7a0$31a0143f@tim>
Message-ID: <000901bef41f$2da7fca0$31a0143f@tim>

[John Machin]
>> I see no ambiguity here; certainly, it is very hard to see anything
>> so ambiguous as to give you the leeway to treat W and H as different
>> from Y! Please elaborate.

[Tim]
> That part isn't ambiguous.  I don't know which person you thought of
> when I wrote "Knuth", but the guy I'm thinking of makes a special
> case of W and H (but not Y) in step 3: ...

On a hunch, I checked:  "the" Soundex algorithm changed between the first and
second editions of Knuth volume 3, and the Wachs example was added in the 2nd
edition to illustrate the W+H elaborations added to step 3.

leaving-it-where-it-started:-even-god-isn't-sure-what-soundex-is-
    supposed-to-do<wink>-ly y'rs  - tim





From kistler at fnmail.com  Thu Aug 19 20:54:36 1999
From: kistler at fnmail.com (Per Kistler)
Date: Fri, 20 Aug 1999 02:54:36 +0200
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu>
Message-ID: <37BCA74C.D3DCF5DD@fnmail.com>

Tom Christiansen wrote:

> The fact that they did icky windows help and not man pages says
> something significant.

Why not create something like perldoc for python?
There must be others missing it...
Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From gmcm at hypernet.com  Fri Aug 20 01:27:48 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 20 Aug 1999 00:27:48 -0500
Subject: Python to Perl Conversions
In-Reply-To: <37bc1303@cs.colorado.edu>
Message-ID: <1276999045-22443458@hypernet.com>

Couple comments -

Tom Christiansen wrote:
>                                 Yes, slices in python don't include
>                                 the termination point.

This might be a surprise if you're not used to it, but it turns out 
to be very nice. Basically any way of expressing a range will result 
in something that includes the left hand side, but excludes the right 
hand. So when you're slicing, chopping up, inserting... you (almost) 
never have to use "i-1".

> delattr(object,name)            n/a?  "delete named attribute from
>                                 object".
>                                 I suppose this might be
>                                         delete $obj->{name}
>                                 but that's pretty uncool to do it
>                                 without asking the object.

I'm not sure I've *ever* seen this used in Python. But it hints at a 
very nice piece of syntactic sugaring you can do.

 class A:
    pass

 a = A()
 a.name = "Joe"
 a.shoesize = 13

All I'm doing is manipulating a dict (a's __dict__), but in a way 
that (sometimes) yields much more readable code than:
 a = {}
 a["name"] = "Joe"
 a["shoesize"] = 13


> reload(module)                  First delete from %INC, then require
>                                 the
>                                 module again, as in:
>                                     delete $INC{"Module.pm"}
>                                     require Module;
>                                 Note that Python doesn't directly
>                                 support nested modules as in
>                                     require
>                                     Dir1::Dir2::Dir3::Module;
>                                 except via a disgustingly named
>                                 hackaround module.

Are you referring to "ni"? Again, this is something that has changed. 
Although Python packages may well have their own "gotchas" for Perl 
folks.
 
> str(x)                          I thought this was the same as
>                                 repr().

It often is, and defaults to repr. But if they are differnt, str
means "as something a user should see" and repr means "as something
a programmer should see" (and, if at all possible, something that
eval will do the inverse of).

> tuple(seq)                      n/a.  This is a nasty hack to deal
>                                 with the
>                                 fact that (1,2,3) is different from
>                                 [1,2,3]. Hm... maybe it's like this:
>                                     $aref = [1,2,3]; 
>                                     @list = @$aref;

When you get used to them, you find that you use lists and tuples 
completely differently. The first thing you find is that because 
tuples are immutable, they can be used as keys in dicts.  But 
more to the point, lists are used as collections, while tuples are 
used as structs or instances of simplistic classes might be. So you 
soon think of them as completely different beasts, even if they are 
syntactically similar.

> type(obj)                       ref($obj) returns its string type

Nit: In Python, "type" returns a type object.
 
> sys.exit(status)                exit(status) and no import is bloody
>                                 necessary just to leave the program.
>                                 But that Python does this by raising
>                                 a SystemExit exception.  Perl can
>                                 use END{} handlers to catch these
>                                 and do at-exit processing, and
>                                 Python doesn't. You have to roll
>                                 your own.

This is not the normal way to leave a Python program, either. The
normal way is just to fall out. Also note that an import of a
built-in costs (almost) nothing.  Namespaces are kept as clean as
possible - maybe because you can clobber any name except a keyword.

> sys.setcheckinterval(reps)      n/a.  This is some event hook
>                                 processing
>                                 thing.

Actually, it's how may byte code instructions to do in one 
"time-slice" if you're using threads.

> string.splitfields(s,sep)       @a = split(/sep/, $s)   # no import
>                                 If you don't want regex magic chars:
>                                 @a = split(/\Qsep/, $s) 

Deprecated. Use the (newer) optional args to split().

> string.joinfields(x, sep)       $s = join($sep, @x);    # no import

Deprecated. Use optional args to join().

> posix.environ                           The %ENV hash; no import
>                                         needed in Perl.  However, assignment doesn't 
>                                         appear to propagate to
>                                         unborn children as it does in Perl.

Use putenv() to do that.


- Gordon



From tchrist at jhereg.perl.com  Fri Aug 20 03:04:43 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Fri, 20 Aug 1999 01:04:43 -0600
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer) 
In-Reply-To: "Tim Peters"'s message of Fri, 20 Aug 1999 02:47:23 -0400 .
             <001401beead7$db473de0$712d2399@tim> 
Message-ID: <199908200704.BAA18741@jhereg.perl.com>

>That cuts both ways, of course, as does any difference.  That is, I once got
>miffed at Perl because my named, nested functions weren't working.  "man
>pages" or not <wink>, it wasn't until I studied the source that I figured
>out Perl effectively moves all named functions to the top level, and if you
>have two nested functions in a package that happen to have the same name,
>the first gets tossed into the bit bucket:

You didn't localize your function!  Remember, in Perl things
are only local if you say they are.

>$x = 5;

>sub outer1 {
>    my $x = 50;
>    sub inner {
>        print "in the first inner, and here's x: $x\n";
>    }
>    &inner();
>}

>sub outer2 {
>    my $x = 500;
>    sub inner {
>        print "in the second inner, and here's x: $x\n";
>    }
>    &inner();
>}

>&outer1();

That could have been like this:

    sub outer1 {
	my $x = 50;
	local *inner = sub {
	    print "in the first inner, and here's x: $x\n";
	};
	&inner();
    }

    sub outer2 {
	my $x = 500;
	local *inner = sub {
	    print "in the second inner, and here's x: $x\n";
	};
	&inner();
    }

    &outer1();

And run just fine.  BTW, you should have gotten a bucket of frightened
warnings out of the compiler about that.  Syntax checking the compile,
with warnings (somewhat like gcc -Wall -pedantic, but without *.o
generation):

    % perl -wc /tmp/timstuff
    Variable "$x" will not stay shared at /tmp/timstuff line 6.
    Variable "$x" will not stay shared at /tmp/timstuff line 14.
    Subroutine inner redefined at /tmp/timstuff line 13.
    Name "main::x" used only once: possible typo at /tmp/timstuff line 1.
    /tmp/timstuff syntax OK

If that is mysterious, bump up the verbosity:

    % perl -Mdiagnostics -wc /tmp/timstuff

    Variable "$x" will not stay shared at /tmp/timstuff line 6 (#1)
	
	(W) An inner (nested) named subroutine is referencing a lexical
	variable defined in an outer subroutine.

	When the inner subroutine is called, it will probably see the
	value of the outer subroutine's variable as it was before and
	during the *first* call to the outer subroutine; in this case,
	after the first call to the outer subroutine is complete, the
	inner and outer subroutines will no longer share a common value
	for the variable.  In other words, the variable will no longer
	be shared.

	Furthermore, if the outer subroutine is anonymous and references
	a lexical variable outside itself, then the outer and inner
	subroutines will never share the given variable.

	This problem can usually be solved by making the inner subroutine
	anonymous, using the sub {} syntax.  When inner anonymous subs
	that reference variables in outer subroutines are called or
	referenced, they are automatically rebound to the current values
	of such variables.
	
    Variable "$x" will not stay shared at /tmp/timstuff line 14 (#1)

    Subroutine inner redefined at /tmp/timstuff line 13 (#2)
	
	(W) You redefined a subroutine.  To suppress this warning, say
	
	    {
	    local $^W = 0;
	    eval "sub name { ... }";
	    }
	
    Name "main::x" used only once: possible typo at /tmp/timstuff line 1 (#3)

	(W) Typographical errors often show up as unique variable names.
	If you had a good reason for having a unique name, then just
	mention it again somehow to suppress the message.  The use vars
	pragma is provided for just this purpose.
	
    /tmp/timstuff syntax OK

Yes, this local function thing is in the Perl Cookbook in the subroutines
chapter.  Larry said that the next release of Perl *shall support

    sub outer1 {
	my $x = 50;
	my sub inner {
	    print "in the first inner, and here's x: $x\n";
	}
	&inner();
    }

    sub outer2 {
	my $x = 500;
	my sub inner {
	    print "in the second inner, and here's x: $x\n";
	}
	&inner();
    }

    &outer1();

But this is lexical scoping, not dynamic scoping.  With the local
sub inner, inner could call fred, and then fred could still
see inner to call it again in indirect recursion.  That couldn't
happen with my sub inner, which is lexically scoped.

BTW, see what I mean about the importance of good error messages?
I'm used to something more than Ken Thompson's "?" retort.  Everytime I
get NameError from Python, I just want to kick it.

And yes, those were compile-time warnigns.  In the developer
release of Perl, you can promote some or all of those into 
fatals if you want, which will make Guido carp less.

    use warnings qw/FATAL uninitialized numeric/;

is more in the "python" style.  And you can always say

    use warnings qw/FATAL all/;

if you feel all warnings should kill you.  Note that that pragma, like use
strict, is lexically scoped to the surrounding nested scope.  That way
one block can selectively enable classes of warnings.  Warning classes
are: 

    all ambiguous closed closure default deprecated exec io misc newline
    numeric octal once parenthesis pipe precedence printf recursion
    redefine reserved semicolon signal substr syntax taint uninitialized
    unopened unsafe untie utf8 void

And you can do

    use warnings;
    no warnings qw/once exec ambiguous/;

or whatever.

You know, it occurs to me that it might be nice to be able to selectively
promote only compile-time warnings to fatals.  Perl does a lot of
compile-time stuff (including constant expression folding, inlining of
some subroutines, and folding of those results!).

What does Python do at compile-time?

--tom



From fredrik at pythonware.com  Mon Aug  9 10:37:01 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 9 Aug 1999 14:37:01 GMT
Subject: ANN: PIL 1.0 final
References: <008e01bedfec$7ffe7110$f29b12c2@secret.pythonware.com> <37aadb42.108851580@news.u-psud.fr> <m37ln9i4or.fsf@schwinger.harvard.edu> <018001bee0e2$14502520$f29b12c2@secret.pythonware.com> <x887ln5xhcn.fsf@rpppc2.hns.com>
Message-ID: <003501bee274$ebe1a570$f29b12c2@secret.pythonware.com>

nbecker at fred.net wrote:
> You need to add -L/usr/X11R6/lib -lX11 to the final link, if you're
> using Linux.

I'm offically on vacation, and my brain is pretty much
switched off, but doesn't that gives you multiple copies
of the Tcl/Tk libraries (one in PIL and one in _tkinter?).

I'd definitely recommend using shared versions of Tcl
and Tk instead...

</F>




From godzilla at netmeg.net  Sat Aug 14 18:29:58 1999
From: godzilla at netmeg.net (Les Schaffer)
Date: Sat, 14 Aug 1999 22:29:58 GMT
Subject: Why use comp.lang.python when you've got comp.lang.perl.misc?
References: <37B5CB95.3FCF571F@appliedbiometrics.com> <199908142005.OAA23215@jhereg.perl.com>
Message-ID: <m27lmyj8nv.fsf@netmeg.net>

> After reading through the _Programming Python_ book, I thought to
> summarize some of my observatations, but the current climate is just
> too hot. 

hey, why should the well behaved among us have to wait for
flame-throwers to chill out. we're here to learn about the languages
we use. bring on your insights, brother.

les schaffer
-- 
____        Les Schaffer              ___| --->> Engineering R&D <<---
Theoretical & Applied Mechanics          |  Designspring, Inc. 
Center for Radiophysics & Space Research |  http://www.designspring.com/
Cornell Univ.  schaffer at tam.cornell.edu  |  les at designspring.com



From stephan at pcrm.win.tue.nl  Fri Aug  6 03:43:54 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 06 Aug 1999 09:43:54 +0200
Subject: Python and Red Hat Linux 6.0
References: <379CCAC0.AD5D1C2F@postoffice.pacbell.net>
Message-ID: <m34sidpd51.fsf@pcrm.win.tue.nl>

Rob <vaton at postoffice.pacbell.net> writes:

> When programming in Perl or Python, I am using a header like the
> following on an executable file to cause the shell (bash or csh) to run
> the scripts as Perl/Python files:
> #!/usr/bin/perl   for Perl   and #!/usr/bin/python   for Python. I've
> checked and the compilers are located in the appropriate directories. I
> keep getting this when I run scripts:
> bash: <script name>: command not found. Does Linux support the '#!'
> notation in the file header, or do I always have to run the scripts by
> explicitly typing in 'perl' or 'python' before the script name?

Did you do `chmod +x' on the file?
Did you start it with `./script', or did you start it with `script'?

On RH, the current directory `.' is not in your path by default
for security reasons, so you have to mention the path explicitely.

BTW, it's not the shell that interprets the `#!...' line.
It's the kernel. (If it was just the shell, it would be just as
useless as those braindead Windows "shortcuts", sorry for the rant.) 

Greetings,

Stephan



From nospam-orawnzva at xntv.pbz  Mon Aug  2 05:38:11 1999
From: nospam-orawnzva at xntv.pbz (Benjamin Kowarsch)
Date: Mon, 02 Aug 1999 18:38:11 +0900
Subject: Mac Install
References: <37A2A5A9.48D17D9@gci.net>
Message-ID: <nospam-orawnzva-0208991838110001@ppp017-max03.twics.com>

StuffIt is the dominating archiving utility. The installer you got there
is a StuffIt archive, therefore you need to have StuffIt on your Mac, to
expand the archive. Stuffit comes in three flavors: StuffIt Deluxe
(commercial product), Stuffit Lite (shareware) and StuffIt Expander
(shareware/freeware - extractor only). You can donload SuffIt from major
download sites of directly from its producers: Aladdin Systems at
http://www.aladdinsys.com

In article <37A2A5A9.48D17D9 at gci.net>, amosg at gci.net wrote:

> I'm having a problem with the Mac install.  I downloaded the file Python
> 151.sit which expanded to 'Python 1.5.1 Installer'.  When I try to run
> the installer file,  it attempts to run Stuffit.  I did a 'get info' on
> the installer and see that it's kind is 'Stuffit document'.  Seems a
> little weird to me.  Any ideas anyone?  Any help is appreciated.
> 
> Amos

-- 
As an anti-spam measure I have scrambled my email address here.
Remove "nospam-" and ROT13 to obtain my email address in clear text.



From skaller at maxtal.com.au  Sun Aug 15 13:18:09 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sun, 15 Aug 1999 17:18:09 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <m3vham44g5.fsf@atrus.jesus.cam.ac.uk> <14257.54137.72005.705719@anthem.cnri.reston.va.us>
Message-ID: <37b6f1c2.347828751@news.triode.net.au>

On Wed, 11 Aug 1999 15:48:09 -0400 (EDT), "Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> wrote:

>At IPC7, JimH reported on some experiments he'd been doing
><http://www.jpython.org/jpython-talk-1.ppt> with whole program
>analysis in the jpythonc tool.  He reported some impressive results,

	I'm very encouraged by this.

	
John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From skaller at maxtal.com.au  Sun Aug 29 01:48:36 1999
From: skaller at maxtal.com.au (John Skaller)
Date: Sun, 29 Aug 1999 15:48:36 +1000
Subject: Python Type System: An idea for unification
In-Reply-To: <1276273608-41434292@hypernet.com>
References: <37c72e93.1411973891@news.triode.net.au>
Message-ID: <3.0.6.32.19990829154836.009664b0@mail.triode.net.au>

At 10:05 28/08/99 -0500, Gordon McMillan wrote:
>John (Max) Skaller wrote:
>
> [Stuff interesting enough to perhaps revive the Types SIG...]

	.. and some of it is actually implemented!

> [Briefly, and subject to my misinterpretation, it appears that
>  Viper objects always have a type obj; said type obj may be
>  a class instance; said type obj is appended to the normal
>  attribute search path]

	"Usually" the type object will be a class, rather than
an instance.

>>  class PyIntType:
>>   def succ(x): return x + 1
>>   def pred(x): return x - 1
>>
>> To use this we try out the Viper code:
>> 
>>  x = 1
>>  print x . succ()
>
>At first glance, these are class methods. But not really, because the 
>bound method object binds the explicit arg 'x' to the user object (in 
>this case also named 'x'). 

	I'm confused!

	1) the function 'succ' REALLY is an ordinary function.
	2) the class PyIntType REALLY is an ordinary class
	3) the object

	(1).succ

	REALLY is an ordinary bound method.

There is NO difference here from Python 1.5.2, it is ONLY the
lookup that is different. [But see below, even the lookup
is exactly what Python does right now!] Try this:

	print [].append

and you will get 

	"<built-in method apppend of list object at 808888>"

so that you can see even in Python 1.5.2, bound methods
do not have to have an 'im_self' object which is a class
instance.

	What IS somewhat strange, however is that
'instances' of the TYPE PyIntType are integers,
NOT class instances: an object of CLASS PyIntType:

	object = PyIntType()

is a class instance, it has type Instance, class PyIntType.
Whereas an integer has type PyIntType, and it doesn't
have a class because it is not a class instance.
This is, however, no different from python 1.5.2!

>But that's not

> class PyIntType:
>  def succ(self): return self + 1
>  ...

	Yes, this is exactly equivalent!

>because "self" in this case would be the instance of PyIntType, 

	No: what you mean I think is that

	object = PyIntType()

does not have type integer, and if you call succ:

	object.succ()

WOOPS: an error because PyIntType does not have an __add__ method.

>not the user object (which "has a" instance of PyIntType). So _something_ 
>is special here. 

	Correct. The automatic searching of the type
of an object is vaguely special (but not as special
as you think!!)

	However, everything else is stock standard python;
although it is also true that the 'type' of an object
no longer has a type whose type's type is itself
(namely TypeType).

	That is, Python 1.5.2's _internal_ TypeXXXX
objects have been dispensed with, and replaced
by arbitrary objects.

	Note that the normal python TypeXXX object
have method tables, and python normally looks them
up .. so what I am doing is not really that different
to Python 1.5.2!

>Ah, you are actually re-binding (the bound to 
>type(x) method) "succ" to the user object ("1"). Right?

	Yes. This is also what Python 1.5.2 would do,
except the method would come from a special,
C constructed TypeXXXX object.

>>  ListOfInt = ListOf(IntType)
>
>So ListOfInt is an instance of ListOf. 

	Yes.

>Unbound methods defined in 
>ListOf are bound methods in ListOfInt. 

	Technically, this doesn't make sense: there are
no 'methods', even in Python, in a class, or even
in an instance. A 'method' is created only during
lookup. So, when a lookup is done on an instance,
a function found in the instances class becomes
a bound method, bound to the instance:

	class X:
		def f(x): pass
	x = X()
	g = x.f

Here, f is a function. It is NOT a method, there is no
such thing as a method (except 'in the abstract').
Thus X.__dict__.f is a function, and an ordinary one.

	However, g = x.f is a bound method: a pair
consisting of an object and a function.

	Python 1.5.2 does not require anything
special about the function, nor the object.
The object can be anything, including a list,
and the function can be any function.
(Technically, any code object I think)

>Presumably, they will be 
>rebound to the object using ListOfInt as its type object.

	Yes, but there are TWO bindings. First,
when a CLASS lookup is done for the instance ListOfInt,
a function 'f' in the class dictionary of ListOf
will be bound to the instance ListOfInt. So we have a bound
method:

	<ListOfInt, f>

here represented by a pairing notation. Now, since that
is a callable function, the lookup on an object x will
convert the callable to a bound method a second time:

	<x, <ListOfInt, f>>

A doubly bound method. 

>Is "new_object" a builtin?

	It will need to be, yes.

>One of us is missing something here! There doesn't seem to be anyway 
>to differentiate between:
>
> class ListOf:
>  def method1(arg1, arg2): ...
>
>being called from the user object x as
> x.method1()   #generic
>  and/or
> x.method1(somearg) #non generic

	What type is 'x'?

>It seems to me you either need an explicit way of marking them as 
>different when writing ListOf, or (I think more generally), writers 
>of type classes need to always handle 2 "self"s:

	Yes. There will always be two selfs. A better naming
convention is:

	the first one is called 'metaself'
	the second one is called 'self'

With that convention, 'self' refers to the object,
and meta-self refers to the type of the object.
Remember that in a generic type constructed as I have
shown, the methods of the meta-type, in this case
ListOf, ar just that: they're shared by all
the instance types such as ListOf(TypeInt).

So you need to bind one of the arguments to the actual
type object -- the metaself argument -- so that the
method can examine the attributes special to the
particular instance of the generic type.

For example, in:

	m = ListOfInt.method

the m function needs to be bound to ListOfInt,
so, for example, it can find out dynamically
what type it is working with: in the example
I gave, the instance ListOfInt has an 'etype'
attribute which determines the type of element
in the list.

But, the m function needs to be bound to 
the client object -- a list of integers ---
as well, so it can do some work, so

	k = intlist.m

the k function is bound to the object 'intlist'
so it can actually do workk on the object.

> class PyIntType:
>   def succ(self, obj): return obj + 1
>   # "self" is the instance of PyIntType,
>   # "obj" is the "self" of the user object - some obj whose type
>   # object is an instance of PyIntType   
>   ...

	For ordinary types like integers, the 'self'
object is an integer: the PyIntType is not generic,
and the type is a CLASS object, not an INSTANCE
object of the class PyIntType.

	However, you could make a generic integer type,
something like:

	class PyNumberMetaType:
		def succ(metaself, self):
			return metaself.addmethod(self,1)


>Then, of course, you are not rebinding a method; you are binding the 
>2nd arg of an already bound method.

	That is the case when you are using the paradigm:

	class MetaType: ...
		def __init__(args ..)

where the __init__ function constructs an instance type object.

>Cool stuff! 

	Yeah. Just consider: I only examined the cases
where I used a class (for standard object types),
and a class (for meta types). However, 
there is no restriction: the type object can be anything.

	But you know what is REALLY cool?

	This is what Python 1.5.2 does now!
There is only one real difference, and that is that
instead of using a TypeXXX object written in C,
I allow any object, and instead of using the fixed
'vtable' lookup, I just use 'ordinary' lookup.
Note that ordinary Type objects in Python
provide in the vtable, a 'getattr' method
for doing lookups for named attributes.

	Note: Viper doesn't use vtables (or objects).
Instead, the interpreter 'knows' how to do all the
standard operations: the functionality is built
into the interpreter, rather than encapsulated
in a Type object.

	This makes it harder to extend than CPython,
but is necessary for implementing a compiler:
you can't optimise encapsulated methods.

>And you successfully avoided ever using the dreaded 
>"meta" word...

	WOOPS. It seems I didn't. :-)

-------------------------------------------------------
John Skaller    email: skaller at maxtal.com.au
		http://www.maxtal.com.au/~skaller
		phone: 61-2-96600850
		snail: 10/1 Toxteth Rd, Glebe NSW 2037, Australia





From stephen at webadmins.com  Sun Aug 15 01:45:49 1999
From: stephen at webadmins.com (Stephen)
Date: Sun, 15 Aug 1999 01:45:49 -0400 (EDT)
Subject: Python CGI Scripts
In-Reply-To: <7p5gu7$nlq@dfw-ixnews14.ix.netcom.com>
Message-ID: <Pine.LNX.4.04.9908150140180.584-100000@gnr.u2me3.com>

I'm just getting into python really, and being a perler for a couple of years
I've gotten it to do just about everything I've ever needed for SA'ing. As I
thumb through 'Programming Python' it looks like just about everything that can
be done with perl can be done with python.

But... have a lot of people adopted python as their new cgi language? This is
what I do primarily lately and I found only one program in the book called
'cgimail.py'. As I read through the code I notice that the h1 tags, the form,
the header infomation, etc etc are all being 'print'ed manually. Is there no
nifty CGI.pm equivelent in python?

Comments?

/***************************************************************\
 * Stephen ("xinu") Klassen -- Unix Systems Administrator      *
 * Website: http://stephen.webadmins.com/   UIN: 19224641 (wk) *
\***************************************************************/




From aycock at csc.uvic.ca  Tue Aug 10 16:59:41 1999
From: aycock at csc.uvic.ca (John Aycock)
Date: 10 Aug 1999 20:59:41 GMT
Subject: Python parser generators
References: <NDBBIAPHEKGOBIHCBLIIIEDHCAAA.mcfletch@vrtelecom.com>
Message-ID: <7oq3rt$4l92$1@uvaix7e1.comp.UVic.CA>

Mike Fletcher (mcfletch at vrtelecom.com) wrote:
: Shameless Plug:
: http://members.home.com/mcfletch/programming/simpleparse/simpleparse.html

Oh dear, we're into shameless plugs already :-)

Ok, guess I'll add my advertisement too:

	http://csr.uvic.ca/~aycock/python

is a package for implementing little languages.  Includes a
(powerful, slow) parser class.

John



From kernr at mail.ncifcrf.gov  Sun Aug 15 13:12:17 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Sun, 15 Aug 1999 17:12:17 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <934326918snz@vision25.demon.co.uk> <37B4500C.276AD22B@basho.fc.hp.com> <37b6dcae.342432061@news.triode.net.au>
Message-ID: <37b6f43e.1175088@news.erols.com>

On Sun, 15 Aug 1999 15:36:39 GMT, skaller at maxtal.com.au (John (Max)
Skaller) wrote:

[snip]

>This last point is the real killer. I have both a Linux and
>NT box. I have no problem with building C modules
>on Linux, but I don't _have_ a compiler on the NT box that
>will build C modules compatible with the Python implementation
>I have. [I have Borland C++, and gcc on the NT box,
>but not MSC, which was used to build my NT based Python]

You can use GCC 2.95 for mingw32
(http://www.xraylith.wisc.edu/~khan/software/gnu-win32/gcc.html)
to compile Python extensions on NT.  See my page
http://starship.python.net/crew/kernr/mingw32/Notes.html

Such extensions will work with the distributed Python interpreter for
Win32.

[snip]

>John Max Skaller                ph:61-2-96600850              
>mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
>http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From thomas at xs4all.nl  Sun Aug 22 15:01:43 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Sun, 22 Aug 1999 21:01:43 +0200
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer)
In-Reply-To: <001601beeaef$e795f380$712d2399@tim>
References: <199908200704.BAA18741@jhereg.perl.com> <001601beeaef$e795f380$712d2399@tim>
Message-ID: <19990822210143.A706@xs4all.nl>

On Fri, Aug 20, 1999 at 05:39:30AM -0400, Tim Peters wrote:

[Tim about Python's compiler & interpreter]
> It's too busy desperately scrounging for excuses to kill your program with a
> bogus SyntaxError to do anything useful <wink>.  About the only compile-time
> optimization is that function locals (in the absence of "exec" and "import
> *" statements) are mapped to consecutive indices into a contiguous runtime
> vector.
> 
> "exec" and "import *" can alter the *set* of local names, so their presence
> inhibits that optimization.  That's why "exec" is a stmt rather than a

[..]

Out of curiosity, is the compiler smart enough to see that exec statements
like

exec "a = 1 * 10" in {}

or

exec "foo = bar * zed" in {'bar': 5, 'zed', 'z'}

are not able to modify any namespace ? I can understand that something like

exec "foo = bar * zed" in myenv

is too indirect for the compiler to worry about, but the upper two examples
do not use any type of object that isn't 100% certain at compile time... Or
is it possible to redefine builtins like ints, dicts and strings ? ;) (Or
isn't the compiler aware of this ?)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jeremy at cnri.reston.va.us  Thu Aug 19 12:01:09 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 19 Aug 1999 12:01:09 -0400 (EDT)
Subject: whrandom documentation (was Re: Unsorting(randomizing) a sequence)
In-Reply-To: <14268.6727.737269.35392@anthem.cnri.reston.va.us>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<14265.47748.241386.291144@buffalo.fnal.gov>
	<wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
	<m3672dwojr.fsf@pcrm.win.tue.nl>
	<14266.49377.51221.305982@dolphin.mojam.com>
	<14266.59802.56315.731454@bitdiddle.cnri.reston.va.us>
	<14268.6727.737269.35392@anthem.cnri.reston.va.us>
Message-ID: <14268.10821.501898.619187@bitdiddle.cnri.reston.va.us>

>>>>> "BW" == Barry A Warsaw <bwarsaw at cnri.reston.va.us> writes:

>>>>> "JH" == Jeremy Hylton <jeremy at cnri.reston.va.us> writes:

  JH> The intended use of the standard library here is for users to
  JH> import random to get access to whrandom.  whrandom implements a
  JH> specific method for generating random numbers.  It could easily
  JH> be replaced with a different, improved module at some time in
  JH> the future.  If user code imports random, it will automatically
  JH> get the benefit of the new implementation.

  BW> In that case, shouldn't the documentation for the whrandom
  BW> module recommend this usage?

  BW>     http://www.python.org/doc/current/lib/module-whrandom.html

Yes.  I module think it makes more sense to put documentation for
methods like random.random in the random module's documentation.  Then 
whrandom could just explain how it is implemented and point people to
the random module docs.

Jeremy



From kuncej at mail.conservation.state.mo.us  Fri Aug 13 10:33:02 1999
From: kuncej at mail.conservation.state.mo.us (Jeffrey Kunce)
Date: Fri, 13 Aug 1999 09:33:02 -0500
Subject: portable file locking?
Message-ID: <s7b3e671.046@mail.conservation.state.mo.us>

>Is there a single file-locking mechanism that will work on NT and Unix?  The
>fcntl module doesn't seem to be in the NT distribution.

I took a stab at this, extending posixfile to work for NT as well as Unix.
My module seems to work, but I have never tested it thoroughly.

ntposixfile.py at http://starship.python.net/~jjkunce/ 

  --Jeff





From mikael_aronsson at mail.bip.net  Fri Aug  6 15:28:14 1999
From: mikael_aronsson at mail.bip.net (Mikael Aronsson)
Date: Fri, 6 Aug 1999 21:28:14 +0200
Subject: __init__ question
Message-ID: <000101bee041$d5d51440$43f9143e@pc58-mikael.fas.flygt.com>

Hi !

I have a small Python class called Vector, it is a simple 3D vector class
with an x,y and z variable.

I would like to have multiple constructors for this (like you can in C++).

Example:

a = Vector()  # x=0 y=0 z=0
a = Vector( 1)  # x=1 y=1 z=1
a = Vector( 1, 2, 3)  # x=1 y=2 z=3

Can this be done in Python, or can I only have one __init__ method  ?

Mikael






From MHammond at skippinet.com.au  Tue Aug  3 08:15:49 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Tue, 3 Aug 1999 22:15:49 +1000
Subject: stuck on COM interface to Word
References: <01BEDD9B.6052EFB0@exchange1.geometria.hu> <37a6d12b.192256@news.mch.sni.de>
Message-ID: <7o6min$28v$1@m2.c2.telstra-mm.net.au>

Nikolai Kirsebom wrote in message <37a6d12b.192256 at news.mch.sni.de>...
>In Arpad's example on *Last print date*, the returned <time object> is
>converted to an integer using the function *int*.  On my installation
>I get the error: 'ValueError: illegal value'.

This (probably) means the date is before 1970, the Python time module epoch.

>Should the *int* function be able to convert the <time object> and if
>so - is there something I've missed out ?

You can either use the builtin methods of the time object (documented in the
help file) or if you want to use dates seriously, get mxDateTime, which has
a function for converting from a COM date.

Mark.






From mcfletch at vrtelecom.com  Mon Aug  9 00:02:13 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Mon, 9 Aug 1999 04:02:13 GMT
Subject: Python parser generators
In-Reply-To: <934154252snz@vision25.demon.co.uk>
References: <934154252snz@vision25.demon.co.uk>
Message-ID: <NDBBIAPHEKGOBIHCBLIIIEDHCAAA.mcfletch@vrtelecom.com>

Shameless Plug:

http://members.home.com/mcfletch/programming/simpleparse/simpleparse.html

It is primarily targeted at writing modestly fast parsers for mid-sized
files using ebnf declarations.  Creates parse trees (as opposed to running
code in-line).  I have a slightly more advanced version which I should be
posting within a few weeks (as soon as I have a chance to do an example for
the new feature (allows adding arbitrary mxTextTools tables to the parsing
definition)).

Is released under a Python-style license.

Enjoy yourself,
Mike




From m.faassen at vet.uu.nl  Wed Aug  4 07:51:36 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 4 Aug 1999 11:51:36 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com>
Message-ID: <7o99g8$72o$1@newshost.accu.uu.nl>

Nigel Head <nhead at houbits.com> wrote:
> Ian Clarke wrote:

>> I am presently working on a CGI based system in Python, and the
>> experience gave me an idea for a simple Python program, I would
>> appreciate any comments (such as "don't bother, it's been done" or
>> "sure, I have loads of spare time, I will do it and release it under the
>> GPL").

> I'm not going to say either of those! I have done it, although I'm not
> terribly happy with the results so far and it's on the back burner for
> the moment .... I need to figure out if it's still useful in a Zope
> world - which means I have to learn and understand Zope!

In the Zope world, the idea is to separate layout/HTML from programming
logic as much as possible so that both can be developed and maintained
independently. For HTML and such Zope has DTML. DTML is a simple language
of tags that can be embedded in HTML, so you can do for instance:
(Zope2 syntax)

<p>My web page</p>
<dtml-if foo>
<p>Foo too!</p>
</dtml-if>

You can also use Python expressions inside DTML, which is sometimes
very handy. These Python expressions are secure; they have to be so,
because you can change DTML from a web management interface, and a normal
Zope user shouldn't be able to screw up the server or the Zope database.

For instance you can loop over a sequence like this:

<ul>
<dtml-in "[1, 2, 3, 4, 5]">
<li><dtml-var sequence-item></li>
</dtml-in>
</ul>

Zope also allows you to extend it using Python. The disadvantage is that
currently can't happen through the web management interface -- a Zope
'external method' (in Python) can do anything at all to the Zope database
and is not restricted by security, so that it's not a good idea to let
it be editable through the web.

[snip]
> Also I haven't yet dealt with putting the python into rexec for execution-
> but this is only needed if a web master wants to give their clients access
> to
> these facilities as opposed to doing things themselves.

There has been some vague idea floating around in the Zope community to
come up with a Python Method that *can* be edited through Zope's web
management interface. It should have the right security machinery, of course.
The Zope philosophy is to keep Python code and HTML separate as much as
possible, but even then a *programmer* will find Python methods very
useful, and in special cases it'll be useful to mix them all anyway.

So it might be nice if you folks figured out a way to include this kind
of functionality in Zope. That would at least be ideal for _me_, as I use
Zope. The tricky bit would be delving deep into Zope to make it all work. :)

Regards,

Martijn




From thomas at xs4all.nl  Fri Aug 27 11:57:14 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Fri, 27 Aug 1999 17:57:14 +0200
Subject: how to get output from system call into data object
In-Reply-To: <863dx5gmrr.fsf@tocol.chc.us.ac.com>
References: <863dx5gmrr.fsf@tocol.chc.us.ac.com>
Message-ID: <19990827175713.E28043@xs4all.nl>

On Fri, Aug 27, 1999 at 12:24:08PM -0500, Stephen M. Shelly wrote:


> I know this is not what I should be doing, but I do not have a 
> good enough understanding of python to figure out how to create 
> my list of data objects 
> from this output.
> 
> is it like:
> MyList = system("Command | awk {print what i need}") ??????????

Well, it depends on what kind of list you want. Does the awk print one line
for each row in your list, and are columns seperated by spaces ? Then this
should work:

import os, string

endlist = []
program = os.popen("command | awk { print what i need}", "r"):

for line in program.readlines():
	endlist.append(string.split(line))


You need to use os.popen instead of os.system, as you want to capture the
output. os.popen returns a file-like object that you can use to read the
lines. Note that the 'readlines()' reads all lines at once, and so doesn't
return until the command you opened with popen() is finished. If the command
| awk {etc} is going to take a long time, and you want to see that it's
working, you may want to do something like

while 1:
	line = program.readline()
	if not line:
		break
	endlist.append(string.split(line))
	print string.join(endlist[-1])

instead of the for loop above.

If your data isn't seperated by whitespace but by some other character, you
can give this as the second character to split:

>>> string.split("a:b:c:d:e:f",":")
['a', 'b', 'c', 'd', 'e', 'f']

Or if the data is fixed-length, you can use slicing instead:

endlist.append( [ line[:5],line[5:10], line[10:] ] )

See the tutorial (http://www.python.org/doc/tut/) for more info on these
topics, plus a LOT more examples. (and better ones ;-)

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From tdixon.no at spam.fwi.com  Wed Aug 25 17:28:13 1999
From: tdixon.no at spam.fwi.com (Timothy Dixon)
Date: Wed, 25 Aug 1999 21:28:13 GMT
Subject: PyKDE for Win ?
References: <7q1als$id9$1@ash.prod.itd.earthlink.net>
Message-ID: <37c45fb2.185911967@midpoint>

On 25 Aug 1999 17:55:08 GMT, kikutani at sprintmail.com (Kikutani Makoto)
wrote:

>
>Looks like a ridiculous qestion, but is there any plan to
>port PyKDE to Win world ?
>
>

Um.... Unless I'm wrong (which has happened a nonzero number of
times), PyKDE is bindings to the KDE Window Manager in XFree86.  Since
KDE doesn't exist in the Windows world, what good would PyKDE do?




From mwh21 at cam.ac.uk  Sun Aug 15 18:30:45 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 15 Aug 1999 23:30:45 +0100
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <m3vham44g5.fsf@atrus.jesus.cam.ac.uk> <37b6e9c3.345781227@news.triode.net.au> <m3pv0o3dc2.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <m3ogg83cdm.fsf@atrus.jesus.cam.ac.uk>

Michael Hudson <mwh21 at cam.ac.uk> writes:

> Not something I've done, but it's always struck me that
> compilers/parsers that only have to deal with good code must be far
> easier to right...

I even go and spell check this post, then I go and write (!) such
garbage at the end of it.

no-command-of-the-English-language-here-ly y'rs - michael



From dwelton at linuxcare.com  Tue Aug 17 13:56:12 1999
From: dwelton at linuxcare.com (David N. Welton)
Date: 17 Aug 1999 10:56:12 -0700
Subject: Stack of C functions created from Python
Message-ID: <87907a8f5v.fsf@linuxcare.com>

I don't really know what the point of this might be, but I thought it
was sort of an interesting bit of code.  I guess it might be useful if
you had to execute a lot of C code that you weren't quite sure about
at run time, and wanted to do it in one big bunch.

/* 
   stackexe.c, by David N. Welton <davidw at prosa.it>

   To compile for Linux:
   gcc -shared -Wl,-soname,stackexe.so -o stackexe.so stackexe.c  
   */

#include <python1.5/Python.h>

typedef struct node {
    int (* func)();
    struct node *next;
} node;

node *head;
node *cur;

static int pa()
{
    puts("a");
}

static int pb()
{
    puts("b");
}

static PyObject *addpa(PyObject *self, PyObject *args)
{
    cur->func = pa;
    cur->next = malloc(sizeof(node));
    cur = cur->next;
    cur->next = NULL;
    return Py_None;
}

static PyObject *addpb(PyObject *self, PyObject *args)
{
    cur->func = pb;
    cur->next = malloc(sizeof(node));
    cur = cur->next; 
    cur->next = NULL;   
    return Py_None;
}

static PyObject *run(PyObject *self, PyObject *args)
{
    cur = head;
    while(cur->next != NULL)
    {
	(*cur->func)();
	cur = cur->next;
    }
    return Py_None;
}

static PyObject *freestack(PyObject *self, PyObject *args)
{
    node *nxt;
    cur = head->next;
    while(cur->next != NULL)
    {
	nxt = cur->next;
	free(cur);
	cur = nxt;
    }
    head->next = NULL;
    return Py_None;
}

static struct PyMethodDef stackexemethods[] = {
    {"addpa", addpa, METH_VARARGS},
    {"addpb", addpb, METH_VARARGS},
    {"run", run, METH_VARARGS},
    {"free", freestack, METH_VARARGS},
    {NULL, NULL}
};

void initstackexe()
{
    Py_InitModule("stackexe", stackexemethods);
    head = malloc(sizeof(node));
    head->next = NULL;
    cur = head;
}

An 'args' field could be added to the node struct, so that you could
pass things to the C functions that way.

Ciao,
-- 
David N. Welton, Web Engineer, Linuxcare, Inc.
415.354.4878 x241 tel, 415.701.7457 fax
dwelton at linuxcare.com, http://www.linuxcare.com/
Linuxcare. At the center of Linux.



From skaller at maxtal.com.au  Tue Aug 10 16:37:40 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Tue, 10 Aug 1999 20:37:40 GMT
Subject: RFC: Viper:  yet another python implementation
Message-ID: <37b075ae.2367293988@news.triode.net.au>

This is a request for comments on Viper, Yet Another implementation
of Python. I'm building Viperi, the interpreter,  and I'm interested in
comments on three issues:

	1) compatability with Python
	2) Extensions
	3) Optimisation

Viperi is being built as a prelude to implementation of an optimising
compiler, Viperc. Unlike previous attempts at a compiler, Viperc will do
type inference, inlining, and special case analysis,
in an attempt to optimise generated code.

The interpreter aims to be compatible with 'pure python':
python not using external C modules, and not fiddling
implementation details too much. Some of the fiddles
will work, and some will not. 

Compatibility:
	1) exceptions: both string and class exceptions are supported,
	the class exceptions are client defined in exceptions.py

	2) rebinding module variables after importing is not permitted
	3) None is a keyword.
	4) range is a keyword

	5) What restrictions can you live with?

Extensions:
	1) full slicing semantics
	2) optional values in dictionaries (Defaults to None)
	3) 'in' applies to dictionaries 'as if a sequence of keys were used'
	4) lexically scoped functions (including lambdas)
	5) optionally typed function parameters

	6) what do you want?

Optimisations:
	By using type inference on pre-imported modules,
it is possible to perform considerable optimisations.
For example, functions can be inlined, and variables
whose types are known can be replaced by native
types. This will generally work best with Viperc, the compiler.
The optimisations will hopefully include using
a mutable string (internally) so that

	x = x + "at the end"

can be changed into an append provided
'x' is not aliased.

	In addition, even in the interpreter,
various special case constructions can be significantly
optimised. For example:

	for i in range(1,10): print i

is controlled by a native for loop, no sequence
object is constructed, not even a lazy one.

	Generally, compiler optimisations
will go hand in hand with restrictions -- to optimise
something, information is required, and that will
require giving up some dynamism.

	It is difficult to optimise individual
modules. It is a different story to optimise
a whole program, where _every_ call
to a function can be traced. Of course,
this may involve restricting what 'exec' can do,
and it may involve other restrictions
(such as not  changing any module bindings
after loading).


John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From kosullivan at snet.wit.ie  Tue Aug 24 08:49:02 1999
From: kosullivan at snet.wit.ie (Kieran O' Sullivan)
Date: Tue, 24 Aug 1999 13:49:02 +0100 (IST)
Subject: Can't use crypt
In-Reply-To: <m3n1vh74l0.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <Pine.LNX.4.10.9908241345120.23432-100000@snet.wit.ie>

On 24 Aug 1999, Michael Hudson wrote:

> "Kieran O' Sullivan" <kosullivan at snet.wit.ie> writes:
> 
> > When I tyr to compile python 1.5.2 on SuSE Linux 6.0 with the crypt module
> > I get the following error.
> > 
> > make[1]: Entering directory `/Python-1.5.2/Modules'
> > gcc  -Xlinker -export-dynamic python.o \
> >           ../libpython1.5.a   -lieee -ldl  -lm  -o python
> > ../libpython1.5.a(cryptmodule.o): In function `crypt_crypt':
> > /Python-1.5.2/Modules/./cryptmodule.c:21: undefined reference to `crypt'
> > collect2: ld returned 1 exit status
> > make[1]: *** [link] Error 1
> > make[1]: Leaving directory `/Python-1.5.2/Modules'
> > make: *** [python] Error 2
> > 
> > I have compiled python 1.5.2 with crypt on SuSE 5.2 with no problems.
> > There was a previous copy of python 1.5.1 on the SuSE 6.0 machine but
> > python 1.5.2 cpmpiles without any problem when crypt is not included.
> 
> So edit Modules/Setup to add -lcrypt then...
> 
> > When I use -lcrypt instead python will compile but I can't use cthe crypt
> > function.
> 
> Oh. What goes wrong now?

when I try to import crypt I get

ImportError: no module named crypt

This error occurs in all prgrams that use crypt.




From arcege at shore.net  Tue Aug 31 17:41:17 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 31 Aug 1999 21:41:17 GMT
Subject: Python Test Executive
References: <7qh93m$38k$1@nnrp1.deja.com>
Message-ID: <1_Xy3.781$1k5.267672@news.shore.net>

winkjj at my-deja.com wrote:
: I am in the process of using Python as our scripting language for a
: test automation environment.  I have been looking for some sort of test
: executive or test management solution that could help me manage and
: execute many scripts.  While there is lots of software out there that
: can do this, I have not found any that interface to python, or would
: interface a cleanly as I would prefer.  Python has so many features
: that are really beneficial for this type of application, but I have not
: been able to find a tool that would allow me to take advantage of these
: features.

: Has anyone done some work in this area?  I have lots of ideas on how a
: Python based test executive could be implemented, but unfortunately, I
: am stuggling with the build yourself vs. buy off the shelf trade offs
: and really don't have the time to invest in building the tooling from
: the ground up.

: I have also looked at TETWare, which is free on Linux, but $10K for
: WinNT.  Unfortunately, I cannot deploy on Linux (just yet, but I keep
: hoping...), and we must run under Win32.  And once I spend the $10K, I
: still have to build a Python binding, because I have not found one
: yet.  And once thats done, there really isn't a nice GUI interface to
: TET anyways.  So it kind of seems like a waste of 10K.

: Any ideas?

I wrote a Python-based testing program a while back, you might want to
look at that:
  http://www.shore.net/~arcege/python/testsuite.py

This is part of my ExpectPy package, if you wish to see the whole test
subsystem (http://starship.python.net/~arcege/ExpectPy/).  Except where
it is hardcoded to import ExpectPy, it should be pretty platform-
independant.

  -Arcege




From fdrake at acm.org  Mon Aug 23 15:17:38 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 23 Aug 1999 19:17:38 GMT
Subject: RFC: Viper:  yet another python implementation
In-Reply-To: <37bdceb0.797602261@news.triode.net.au>
References: <37b075ae.2367293988@news.triode.net.au>
	<7orbpq$i4s$1@newshost.accu.uu.nl>
	<37b6dea0.342930367@news.triode.net.au>
	<7p7h3p$jto$1@newshost.accu.uu.nl>
	<37bdceb0.797602261@news.triode.net.au>
Message-ID: <14273.40530.536980.526432@weyr.cnri.reston.va.us>

John (Max) Skaller writes:
[regarding the C interface:]
 > 	No, there was an upgrade from 1.5.1 to 1.5.2,
 > 1.6 may be different, 2.0 will certainly be different.

John,
  The C interface should be considered fairly stable.  While there are 
occaisional changes, they tend to be additions rather than revocations 
of the existing functions.
  2.0 should be considered a new beast entirely.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From bgibson at us.ibm.com  Tue Aug 17 16:31:06 1999
From: bgibson at us.ibm.com (bgibson at us.ibm.com)
Date: Tue, 17 Aug 1999 16:31:06 -0400
Subject: (no subject)
Message-ID: <852567D0.00709441.00@d54mta04.raleigh.ibm.com>



  Does anyone have code that will parse RTF input files?

Thanks in advance
Bob





From rbl at hal.cwru.edu  Fri Aug  6 08:21:27 1999
From: rbl at hal.cwru.edu (rbl at hal.cwru.edu)
Date: 6 Aug 1999 12:21:27 GMT
Subject: Help with dictionary, please
Message-ID: <7oek07$so3$1@alexander.INS.CWRU.Edu>

I'm rapidly trying to learn Python and despite reading the documentation
and the "Learning Python" book, I can't figure out how to do the following:

-  I want to have a 2-dimensional array of word counts, indexed by
	a pair of words.  I.e.  dict[word1,word2]
-  In C, I'd just create a 2-dimensional array of integers and increment
	as required, but would have to hash the words into integer indices.

-  I tried the following in Python:
	dict = {}
	prev1 = ''
	prev2 = ''
[snip]

	if prev1 != '' and prev2 != '':
	 	   key = (prev2,prev1)
	           dict[key] = dict[key] + 1
   and got this error:
	  File "Metro:Python 1.5.1:learn2.py", line 34
          dict[key] = dict[key] + 1
                         ^
	SyntaxError: invalid token

I'm obviously missing something here!

Any guidance greatly appreciated!

Thank you,
Rob Lake
rbl at hal.cwru.edu






From phil at ricochet.net  Mon Aug  9 13:50:13 1999
From: phil at ricochet.net (phil)
Date: Mon, 09 Aug 1999 17:50:13 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com> <37AA804D.9A1317C0@houbits.com> <934115482snz@vision25.demon.co.uk> <37ae686e.59511102@news.ricochet.net> <7om63q$cqc$2@newshost.accu.uu.nl>
Message-ID: <37af12de.2492494@news.ricochet.net>

On 9 Aug 1999 09:13:30 GMT, m.faassen at vet.uu.nl (Martijn Faassen)
wrote:

>phil <phil at ricochet.net> wrote:
>[snip discussion of licenses, GPL, etc]
>> I can't forbid commercial use. This was why i chose Python. I have no
>> idea if people can or will  make money on my interface. If Python
>> imposes commercial limitations then i'll have to drop it, pissed-ly,
>> since i spent quite a bit of time writing an extension, but that's the
>> way it goes.
>
>I'm confused. Perhaps you're under the impression Python is distributed
>under the GPL? It's not; it's distributed under its own license which
>(read it) basically allows you to do anything you like with it, commercial
>or not.
<snip very interesting licence discussion>

Thanks. I was confused.  I'd read the Python licence and it was
completely open. The 'GPL' threw me and i was concerned that things
were a-changing.

Less confused-ly  yours,

phil

>
>License-pundit-for-5-minutes-ly yours,
>
>Martijn




From gmcm at hypernet.com  Sat Aug 14 12:14:22 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 14 Aug 1999 11:14:22 -0500
Subject: Enough [Was Re: Why use Perl when we've got Python?!]
In-Reply-To: <slrn7rb18h.cf9.jstevens@bamboo.verinet.com>
Message-ID: <1277478426-31364719@hypernet.com>

John W. Stevens wrote:

[ crap deleted ]

Whatever your contributions to the python community, you're now in my 
killfile, along with a handful of pl hotheads who do the same to us.

- Gordon



From mwh21 at cam.ac.uk  Mon Aug 30 03:51:43 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 30 Aug 1999 08:51:43 +0100
Subject: Question: lists in classes
References: <37CA1803.2FA9261B@ug.cs.su.oz.au>
Message-ID: <m3emgl904w.fsf@atrus.jesus.cam.ac.uk>

Jeff Turner <jturner at ug.cs.su.oz.au> writes:

> Apologies if this has been answered before...

Just a few times, yes. This problem seems hard to categorise and so
difficult to search deja or the FAQ for (is there an entry in the FAQ
for this? I can't find it!).
 
> class foo:
>     list=[]

This makes `list' a *class* variable, not an instance variable.

You want to write:

class foo:
    def __init__ (self):
        self.list=[]

instead. This bites almost everyone at some point I think. It
certainly bit me.

> a=foo()
> b=foo()
> a.list.append(123)
> print b.list
> 
> I would expect the above code to return "[]", since b has not been
> modified since instantiation. However it returns [123].
> 
> How would I tell python that I only want to _define_ the class, not
> instantiate it?

Um, don't really understand what you're asking here. 

To understand what's going on, you should understand that the class
body is executed when the interpreter reads it, not when the class is
created. This can be used to erform some sneaky dynamic hacks, but
most of the time it just does what you want.

you'll-get-used-to-it-ly y'es - Michael



From skaller at maxtal.com.au  Sun Aug 15 11:36:39 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sun, 15 Aug 1999 15:36:39 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <934326918snz@vision25.demon.co.uk> <37B4500C.276AD22B@basho.fc.hp.com>
Message-ID: <37b6dcae.342432061@news.triode.net.au>

On Fri, 13 Aug 1999 11:04:12 -0600, "John W. Stevens" <jstevens at basho.fc.hp.com> wrote:

>Phil Hunt wrote:

>> sorted dictionary loops, eg:

>If this last one is what you want, why not use either my B-Tree
>container class, or my Trie container class?

One of the goals of Viper is too eliminate 
the need for C modules There are several reasons
for this. One is that C modules are hard to implement,
another that the C API changes from CPython version
to version, another is that it isn't really as efficient as
whole program analysis, the API dependent on a _particular_ 
Python implementation (i.e. won't work with JPython),
and finally, it is hard to build a C module, so distributing
Python code including C modules is difficult for
developers. 

This last point is the real killer. I have both a Linux and
NT box. I have no problem with building C modules
on Linux, but I don't _have_ a compiler on the NT box that
will build C modules compatible with the Python implementation
I have. [I have Borland C++, and gcc on the NT box,
but not MSC, which was used to build my NT based Python]

Pure python is easier to distribute. If it could be compiled
to be as fast as C modules, there is no need for the
portability headache.

Having said all that, I think it is worth providing a binding
to existing C modules, but I can't do everything at once,
and the extra work required to do the binding is too much
at the moment.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From nriedlin at my-deja.com  Mon Aug 16 09:25:46 1999
From: nriedlin at my-deja.com (nriedlin at my-deja.com)
Date: Mon, 16 Aug 1999 13:25:46 GMT
Subject: Controlling Python
Message-ID: <7p93ge$2bt$1@nnrp1.deja.com>

Hi all,

in my application (written in C++) I want to use Python as the
scripting language. The application comunicates with the hardware and
when a specific event occurres, a thread is started, in which a python
interpreter is created to execute a certain script. The thread should
have a finer control over the Python-Interpreter as is possible with
PyRun_SimpleFile().
(E.g. I want to be able to stop the thread independent of the
interpreter) I would like to have some kind of "hook" after, say, every
line of python-code.
I thought something like (pseudocode):
while(getline(f, s) {
  PyRun_SimpleString(s);
  my_hook(s, lineno++,...);
}
should work but it fails, whenever the python-code becomes a little
more difficult. E.g. PyRunSimpleString() failes, whenever the simple
string continues in the following line:
if bla:
    pass

Is this the way to go? If yes: how? If no: why not and in which
direction should I turn?

Any hints are most welcom.

TIA

Norbert Riedlin


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From faqtori at hotmail.com  Mon Aug 30 00:27:40 1999
From: faqtori at hotmail.com (Factory)
Date: Mon, 30 Aug 1999 14:27:40 +1000
Subject: Why do we call python a scripting language?
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com> <37c726c4.2148316@news.netset.com>
Message-ID: <MPG.12339a1a18f7e7b09896ed@news.melbpc.org.au>

In article <37c726c4.2148316 at news.netset.com>, someone at someplace.com 
says...
> On 27 Aug 1999 12:59:18 GMT, fredp at multimania.com.nospam (Fred
> Pacquier) wrote:
> But what I find most interesting is that I havent seen much evidence
> that any one uses it.  Is any one using freeze or some derivative or
> work alike?  Most things I find are written for users of python, and
> are just distributed as source.

  Well I am going to be writing a game that uses python embedded in an 
executable, it'll just launch a python script and use that as the main 
program, the low-level stuff will be done in C/C++/asm. The actual script 
will be hidden from the user and the python interpreter will be 
statically linked to the .exe.
  The main annoyance is that the python code is so large, too much crap 
is in the main .dll (erm, speaking of the windows version) that won't get 
used in my game, from about 500k, I reckon I should be able to get it 
down to 200k at least, with packing 100..

  - Factory



From tim_one at email.msn.com  Fri Aug 27 00:06:24 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 27 Aug 1999 00:06:24 -0400
Subject: carpe linguam
In-Reply-To: <m3u2pmcpv8.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <000201bef041$86f81ba0$592d153f@tim>

[Michael Hudson]
[... skipping all the interesting stuff to save time <wink> ...]

> Python isn't perfect; anyone who claims it is is either lying or
> stupid.

Or, in my case, both!

> I much prefer it to perl, but that says as much about me as
> about the languages.

I still use Perl for many a text-crunching one-shot (it's fast and convenient!
can't beat it for that), but spend far too much time helping others track down
their Perl bugs.  Here's one that popped up at work today:

    @a = (1 .. 100);
    $sum = 0;
    foreach (@a) { $sum += $_; }
    print "sum $sum\n";
    print "length ", length(@a), "\n";

That prints:

    sum 5050
    length 3

3?  Oh yes:  the array @a is in scalar context, and in this particular flavor
of scalar context returns its length.  The "length" function then sees 100, and
goes "aha! so how many characters does 100 consume when converted to a decimal
string?".  Three, and "perl -w" doesn't complain.

I've seen that before, so it didn't take long to track down.  But if you're
looking for a good reason to prefer Python for some tasks, it's not a challenge
to find about three of them in that example <0.3 wink>.

do-what-i-don't-mean-ly y'rs  - tim





From OUYFOUF at UYGESR.COM  Wed Aug 11 18:15:50 1999
From: OUYFOUF at UYGESR.COM (OUYFOUF at UYGESR.COM)
Date: Wed, 11 Aug 1999 16:15:50 -0600
Subject: ONLY ADULTS 13085
Message-ID: <110899161550@UYGESR.COM>

Free XXX:

http://207.240.225.250






;E?Sp0=@



From skip at mojam.com  Wed Aug 18 13:42:53 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 18 Aug 1999 12:42:53 -0500 (CDT)
Subject: Wacky Programming Tales
In-Reply-To: <01bee9a0$30018e00$583c08cf@bill.jump.net>
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk>
	<37B54DA3.4210DC10@nospam.com>
	<slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au>
	<wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net>
	<37b62939@cs.colorado.edu>
	<slrn7rccem.ne.bmcd@ripley.sulaco.net>
	<37BAE44B.E66250BD@geo.vu.nl>
	<01bee9a0$30018e00$583c08cf@bill.jump.net>
Message-ID: <14266.61569.441849.181455@dolphin.mojam.com>

    William> if ( something )
    William> {
    William> }
    William> else
    William> {
    William> 	A process that sets the success variable
    William> 	if (! success )
    William> 		goto Error;
    William> }
    William> Error:

    William> Note that there was nothing between the goto and and the label
    William> except the end of the else clause. -Wm

Planning ahead, no doubt. ;-)

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From paul at prescod.net  Fri Aug 27 13:23:14 1999
From: paul at prescod.net (Paul Prescod)
Date: Fri, 27 Aug 1999 17:23:14 GMT
Subject: Slow python on particular computer
Message-ID: <37C6C982.35F2B199@prescod.net>

Has anyone seen a case where Python runs REALLY, painfully slowly on an
otherwise spritely Windows 95 machine? It takes something like 8 seconds
to launch the interpreted and parse a 10 line program. Something must be
mis-configured.

 Paul Prescod



From cjc26 at nospam.cornell.edu  Tue Aug 31 23:09:52 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 1 Sep 1999 03:09:52 GMT
Subject: running python on a floppy?
References: <z6ex3.1541$1i4.128395@typhoon.mbnet.mb.ca> <37CA50EC.4FC03560@home.com> <slrn7sm5m2.k0.cjc26@tankgrrl.bridget.mindriot.net> <slrn7smt7p.geq.wtanksle@dolphin.openprojects.net>
Message-ID: <slrn7sooao.ab.cjc26@tankgrrl.bridget.mindriot.net>

Pada Tue, 31 Aug 1999 06:23:54 GMT, William Tanksley bilang:
| I had understood that the click of death was a har4dware problem in the
| earlier drives, not on the disks themselves.

It can either be caused by a bad disk physically destroying the drive
mechanism, or by a drive going out of alignment or whatever and writing bad
data all over the disk.  In the second case, you can often reformat the
disk and at least reuse it, but there's no way to recover what was on it
before.


| At any rate, if you're
| comparinng Zip disks to floppies, Zips win -- most floppies I've put
| semi-important information on have failed, one just today barely half a
| minute after I placed the info on it.
| 
| I've never had a Zip fail me, ever.

Well lucky you :)
Btw, I've never had a floppy fail on me, ever.

of-course-a-shell-account-with-no-quota-is-the-best-option-of-all-ly yr's, Cliff


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From kistler at fnmail.com  Thu Aug 19 20:57:24 1999
From: kistler at fnmail.com (Per Kistler)
Date: Fri, 20 Aug 1999 02:57:24 +0200
Subject: No man pages? (was Re: More random python observations from a perl 
 programmer)
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu> <19990819123502.A28466@digex.net>
Message-ID: <37BCA7F4.C8C30F18@fnmail.com>

John Landahl wrote:

> What's all this about no Python manpages?
> 
>   $ man python
>   Reformatting page.  Wait... done
> 
>   Misc. Reference Manual Pages               PYTHON(10 April, 1998)

It's there, but I think "perldoc perl" still has more
to say. Why not offer all possible resources?
Even tcl has more...

-Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From fdrake at cnri.reston.va.us  Fri Aug 20 09:31:28 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake, Jr.)
Date: Fri, 20 Aug 1999 13:31:28 GMT
Subject: More random python observations from a perl programmer
In-Reply-To: <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk>
References: <37bc1339@cs.colorado.edu>
	<m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk>
	<37bc4111@cs.colorado.edu>
	<m3n1vne72r.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <14269.22704.714782.33024@weyr.cnri.reston.va.us>

Michael Hudson writes:
 > On this subject I'll just say: If people before you had cared as much
 > about the existence of a python(1) manpage, it would have one.

  People do care about this one, and rightfully so.  Python does have
one on Unix, and has for as long as I can remember knowing about
Python (make your own guess as to how valuable that is).  If "man
python" doesn't bring it up, there's a problem with your installation.
  The changes to the man page which I promised the list yesterday are
now in the CVS repository.  I'll even mail a copy to Tom C. if he asks 
nicely.  ;-)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From viking at removeme.cit.org.by  Sat Aug 14 04:29:21 1999
From: viking at removeme.cit.org.by (Eugene Zaikonnikov)
Date: Sat, 14 Aug 1999 11:29:21 +0300
Subject: take the 20 second lang. survey
References: <37B4B3F0.77E6C59A@netplus.net> <GN3t3.4218$Rh.194251@typ11.nn.bcandid.com>
Message-ID: <934619272.442851@lxms.cit.org.by>

David C. Hoos, Sr. <david.c.hoos.sr at ada95.com> wrote in message
news:GN3t3.4218$Rh.194251 at typ11.nn.bcandid.com...
>
> Luke Loeffler <floods at netplus.net> wrote in message
> news:37B4B3F0.77E6C59A at netplus.net...
> > Take the 20-second language survey.  No personal information must be
> > entered.  This is only a survey steming from my personal interest--it is
> > not affiliated with any organization.
> >
> > http://www.recordingwebsite.com/language.html
>
>
> I didn't answer the survey, because when it came to the questions
> about OOL, the first internationally standardized OOL (i.e., Ada)
> was not the choices offered.

Common Lisp/CLOS isn't listed either - although it is an ANSI standardized
OO language. Moreover, it is not even in the 'your primary language' list.

--
    Eugene.





From bucodi at worldnet.fr  Mon Aug  2 06:07:15 1999
From: bucodi at worldnet.fr (pserrand)
Date: Mon, 02 Aug 1999 12:07:15 +0200
Subject: IntVar - to detect a change in it
References: <7o3k4f$74n$1@nnrp1.deja.com>
Message-ID: <37A56DD3.EFACB39A@worldnet.fr>

Ole-Kristian Villaboe wrote:

> Is there anyway an altering of a variable (Int-, String- or Double-)
> (by the user, via the GUI) could call a function in order to tell the
> environment that it has changed since the initial setting(or a static
> comparison value)
>
> Ole-Kr. Villaboe
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

You could use the lostfocus event of the widget

Pierrette




From lutz at rmi.net  Fri Aug 13 19:16:59 1999
From: lutz at rmi.net (Mark Lutz)
Date: Fri, 13 Aug 1999 23:16:59 GMT
Subject: Python Pocket Ref index available
Message-ID: <199908132316.RAA05367@shell.rmi.net>

In response to reader requests, O'Reilly has just 
posted an index for the Python Pocket Reference, at:

    http://www.ora.com/catalog/pythonpr/

They're either going to sneak it in to the next printing, 
or hold onto it for the 2nd edition.  For now, you can
download and print what's on the page above.

Cheers,

--Mark Lutz



From gerrit.holl at pobox.com  Mon Aug 30 08:55:06 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 30 Aug 1999 12:55:06 GMT
Subject: Upper limit?!
In-Reply-To: <014f01bef2bd$3fd09e90$f29b12c2@secret.pythonware.com>; from Fredrik Lundh on Mon, Aug 30, 1999 at 07:56:59AM +0000
References: <37CA1E0F.FA2A05A3@megsinet.net> <014f01bef2bd$3fd09e90$f29b12c2@secret.pythonware.com>
Message-ID: <19990830145506.A6858@optiplex.palga.uucp>

On Mon, Aug 30, 1999 at 07:56:59AM +0000, Fredrik Lundh wrote:
> David Oppenheimer <davidopp at megsinet.net> wrote:
> > I was playing around with the command line and discovered that when I
> > enter 2 to the 30th power, I get a number back...when I enter 2 to the
> > 31st power I get an overflow error.  Is 2 to the 30th power near the
> > upper limit for computations?!
> 
> just above it, on a 32-bit platform.  try using
> use long integers instead, and compare the
> result with the sys.maxint value:
> 

And are longs slower too?
If not, one can use longs always!

regards,
Gerrit.



From steve_w at localhost.localdomain  Mon Aug 23 22:07:44 1999
From: steve_w at localhost.localdomain (steve_w at localhost.localdomain)
Date: Tue, 24 Aug 1999 02:07:44 GMT
Subject: RPM for Python 1.5.2 ?
References: <slrn7qrn3u.7fj.dyoung@baltazar.baltazar.org> <37ADE4C1.194C31D6@lemburg.com> <19990809113731.A6757@baltazar.baltazar.org> <37AFD5A9.4C45332F@lemburg.com> <7ps5t7$2u6$1@nntp.Stanford.EDU>
Message-ID: <slrn7s3vko.to.steve_w@localhost.localdomain>

On 23 Aug 1999 19:03:03 GMT, Amit Patel <amitp at Xenon.Stanford.EDU> wrote:
>| > On Sun, Aug 08, 1999 at 10:12:49PM +0200, M.-A. Lemburg wrote:
>| > > David Young wrote:
>| > > >
>| > > > Has anyone built an RPM for Python 1.5.2 ? I think it's sort of curious
>| > > > that RedHat hasn't already. Anyway, I'm building my own package right now,
>| > > > but I'm afraid I might be wasting my time....
>| > >
>| > > Try:
>| > >
>| > >       http://www.andrich.net/python/download.html
>
>
>I had good luck with these RPMs on RH 6 except  when I tried
>installing the Tkinter stuff:
>
>  # rpm -Uvh python-_tkinter-1.5.2-2.i386.rpm 
>  error: failed dependencies:
>          blt >= 2.4c is needed by python-_tkinter-1.5.2-2
>          libBLT.so.2 is needed by python-_tkinter-1.5.2-2
>
>I don't know what blt is!  I did erase RH's tkinter package, but it
>didn't help.  Do I need to install Tcl/Tk extensions to make it work
>with Python?  :(  

blt is a set of widgets for TK, so look for a blt RPM wherever your
RH 6.0 distribution keeps TCL/TK RPMS.


Steve W.



From donn at u.washington.edu  Tue Aug 24 12:24:53 1999
From: donn at u.washington.edu (Donn Cave)
Date: 24 Aug 1999 16:24:53 GMT
Subject: Writing data to a pty file descriptor
References: <37C1E10B.D2170157@cisco.com> (none)
Message-ID: <7puh0l$1bna$1@nntp6.u.washington.edu>

Quoth Aaron Rhodes <aarhodes at cisco.com>:
| I'm trying to write data to a pty (pseudo terminal).
| I'm using the pty module (pty.writen) after forking a
| child with a controlling terminal, execing the shell, etc.
| (similar to what Expect does or the code in Ch.19 of
| Stevens' Advanced Programming in the UNIX environment)
|
| I'm not sure why but I can't write certain characters
| to the pty. I've written the following command to the pty to set
| up for a file input from the shell command line.
|
| $ cat > file.uue << EOF
| >
|
| However, as soon as I try to write certain characters,
| the shell seems to interpret them incorrectly.  One
| character that does this is the back-tick -- `
| character.  This seems odd to me since I can input
| the characters from stdin just fine... but not by
| writing to the pty directly.
| Any ideas about why a write to the pty from python
| would fail?

Not really, but in the meantime, I think you should change
"EOF" to "'EOF'".  Quote it, in other words.  That would
seem to make no sense, but it's a cue to the shell that the
"here document" following is not to be subject to expansion
of variables and so forth - including `.  (I don't know, you
don't really say what happens, or what "interpret them
incorrectly" means to you.  The shell is interpreting them
correctly, it's your expectation that's incorrect, but if
you want the backquotes to be interpreted, then my advice
is misplaced.)

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu




From TheBlueWizard at pressroom.com  Thu Aug 19 20:49:51 1999
From: TheBlueWizard at pressroom.com (The Blue Wizard)
Date: Fri, 20 Aug 1999 00:49:51 GMT
Subject: Viewing contents of an object
Message-ID: <01beeaa5$c7f6db60$7f8345c6@TheBlueWizard>

Hi Pythoneers!

I would like to be able to view the content of an arbitrary object, e.g.
view(myObject) would print
class: myClass
class variables:
    counter   = 6
    topicoftheday = "Python is fun!"
instance variables:
    x = 4
    y = 7

Thanks!

Cheers!

The Blue Wizard




From tim_one at email.msn.com  Sun Aug 22 03:11:42 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 22 Aug 1999 03:11:42 -0400
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc29d8@cs.colorado.edu>
Message-ID: <000001beec6d$9753e240$542d2399@tim>

[posted & mailed]

[Tom Christiansen, on "in" and dicts]
> Well, it's the polymorphism thing.  If "+" is supposed to work on
> anything (which I don't agree with, but that's not my decision),

To the contrary, for your objects it *is* your decision whether they
implement semantics for "+"; by default, they won't.  For Python's builtin
types, it was Guido's decision, and most types *don't* implement "+" (for
example, none of function, dict, file, traceback, class, method, module or
type objects).  Polymorphism is a freedom in Python, not a requirement.

> I don't understand why "in" shouldn't.

Neither do we <wink>.  1.6 will probably introdoce a class __contains__
method to let users implement their own "in" semantics.

> I keep thinking that everything should work on everything.

The builtins strive to stick to those type/operator combos that make plain
sense; Guido has never been a fan of gratuitous orthogonality.

> For example, this surprises me:
>
>     x = {1:2, 3:4}
>     y = {10:20, 30:40}
>     z = x + y
>
> The latter isn't legal.

In the early years it was debated, and there were conflicting views about
what to do in case of collisions; so rather than introduce a continuing
source of surprises, the idea was dropped.  Merging dicts proved too useful
too often for that to stand, so a dict.update(otherdict) method was
introduced:  since it's most useful most often to merge in-place, and "x +
y" suggests anything *other* than in-place merging, a method was used
instead.  If += had been implemented first, I imagine dict merge would be
spelled via + instead.

> But in Perl, you can merge hashes easily.

Also in Python, although since there's no split between objects and
references to objects, unlike as in Perl the operation is always spelled the
same way.

> ...
> Of course, that's [in Perl] not super efficient.

It is in Python <wink>.

> And that overright duplicates, rather than making lists of them.
> Still, that's not a bad behaviour.

Ditto, ditto.

> I can't get used to not being able to do hash slices, as in
>
>     @revhash{values %hash} = keys %hash;
>
> Or
>
>     print sort(@hash{@desired})

General gather/scatter functions are short & easy to write, if you want
that.  For example, here's the harder one:

    def scatter(container, indices, values):
        for i in range(len(indices)):
            container[indices[i]] = values[i]
        # and some people like to return the mutated container here,
        # although most don't

Note something interesting here:  because of that dreaded <wink>
polymorphism, with 3 lines of code you've now implemented scattering for
lists, arrays, dicts, and all current & future user-defined sequence &
mapping types too!

What bothers me in the above is the "range(len(indices))" business:  it's
too common to be that wordy.  Parallel sequence iteration would be a good
language addition (to Python and Perl!).

Now that you've seen the obvious way, there's also a fast but obscure way:

    def scatter(contaioer, indices, values):
        import operator
        map(operator.setitem, (container,)*len(indices),
            indices, values)

That does the buik of the work at C speed.

there's-always-some-way-to-do-it<wink>-ly y'rs  - tim





From banderson at boi.hp.com  Fri Aug 27 15:46:34 1999
From: banderson at boi.hp.com (Bill Anderson)
Date: Fri, 27 Aug 1999 13:46:34 -0600
Subject: Python 1.5.2 on HPUX 10.20
References: <37C3D5BD.72024971@abg1.siemens.de> <19990825140816.R28043@xs4all.nl> <37C3F4E2.4953BD78@abg1.siemens.de>
Message-ID: <37C6EB1A.AC8F44E7@boi.hp.com>

Volker Hetzer wrote:
> 
> Thomas Wouters wrote:
> 
> > > Is this serious? What am I to do?
> [info that the test has to do with locking.]
> 
> So I can safely ignore it.
> Thanks a lot!
> Volker


hmm ... would you mind emailing me (in private, of course) your
makefile?

I am trying to get it to compile. It did so once, without threads, and
has failed to do so each and everytime since (even with virgin source??)
w or w/o threads :-(

I'd really like to get threading, since I need to run Zope on it.

-- 
Bill Anderson 					Linux/Unix Administrator, Security Analyst
ESBU (ARC)					banderson at boi.hp.com
My opinions are just that; _my_ opinions.



From matthewm at zebrasoft.co.nz  Sun Aug  1 23:19:41 1999
From: matthewm at zebrasoft.co.nz (Matthew Miller)
Date: Mon, 02 Aug 1999 15:19:41 +1200
Subject: RMI and Python
References: <Pine.GSO.4.10.9908011131490.25758-100000@xmission.xmission.com>
Message-ID: <37A50E4D.C6D7430B@zebrasoft.co.nz>

Howabout JPython - a 100% pure java implementation of python which gives you
full access to all the java packages and let's you create/manipulate, etc java
objects from within a python environment. Check it out at www.jpython.org/

Tim Egbert wrote:

> Java is able to do remote method invocation (RMI) which is a neat concept.
> It is the basis for Jini, another neat idea.  Does Python have anything
> like RMI?
>
> Tim Egbert

--
Matthew Miller
Zebra Software Ltd.

E-mail: matthewm at zebrasoft.co.nz
Ph: 025-2920625 / FAX: 03-3842397





From davidopp at megsinet.net  Mon Aug 30 04:20:14 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Mon, 30 Aug 1999 04:20:14 -0400
Subject: Upper limit?!
References: <37CA1E0F.FA2A05A3@megsinet.net> <014f01bef2bd$3fd09e90$f29b12c2@secret.pythonware.com>
Message-ID: <37CA3EBD.4077AF18@megsinet.net>

Fredrik Lundh said:

"on a 32-bit platform.  try using
use long integers instead, and compare the
result with the sys.maxint value:

>>> 2L**31
2147483648L
>>> import sys
>>> sys.maxint
2147483647

if you upgrade to an alpha, you'll get:

>>> 2**31
2147483648
>>> import sys
>>> sys.maxint
9223372036854775807"

Wow, that's quite a difference! on an Alpha you can have a 19 digit number
as opposed to a 10 digit number on my Pentium.  So the limit isn't actually
a factor of the programming language but a virtue (or limitation) of my
processor.  I wonder if Intel knows about this?!   LOL  Makes me want to
sell this hunk of crap Packard Hell P-75, get rid of Win98 and buy an Alpha
running Linux.   That PSN number thing scares the hell outta me any ways,
I'm either going AMD or Alpha next go around.

David O.




From roy at popmail.med.nyu.edu  Tue Aug  3 15:45:58 1999
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Tue, 03 Aug 1999 15:45:58 -0400
Subject: A question of style
Message-ID: <roy-0308991545580001@qwerky.med.nyu.edu>

I'm using BaseHTTPServer.  I need to maintain a long-lived database
connection (using oracledb), and the only persistant object is my server,
so the db connection is stored there.

I've got a class for each type of page that gets displayed, with widgets
on each page that need to access the database.  Thus, I end up with stuff
like:

self.page.handler.server.db.cursor()

inside my widget class methods.  This works, but it's kind of messy.  I'm
thinking of giving my base widget class a cursor method, which looks like:

def cursor(self):
   return self.page.handler.server.db.cursor()

and then I can just do self.cursor() inside my widgets.

I can't think of any particularly compelling technical reasons why that's
a good or bad thing.  Seems more like a matter of style than anything
else.  Comments?

-- 
Roy Smith <roy at popmail.med.nyu.edu>
New York University School of Medicine




From philh at vision25.demon.co.uk  Mon Aug 30 10:45:58 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 30 Aug 99 14:45:58 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <MPG.1233dfb6145766d998968c@news.mop.no>
Message-ID: <936024358snz@vision25.demon.co.uk>

In article <MPG.1233dfb6145766d998968c at news.mop.no>
           earlybird at mop.no "Alexander Staubo" writes:
> In article <935279776snz at vision25.demon.co.uk>, 
> philh at vision25.demon.co.uk says...
> > 
> > 
> > It looks like version 0.0.1 of Parrot, my text-based GUI builder,
> > will be ready within a day or so. It doesn't actually do anything
> > useful yet; v0.0.1 is a technology demonstration version, which
> > is intended to verify that the basic design is more-or-less
> > sensible. Would anyone like to have a look at it, and possibly 
> > suggest some improvements? If so, tell me, and I'll put it on
> > my website.
> > 
> > Some information about Parrot follows:
> [lots of stuff snipped]
> 
> I see no mention of two-way synchronization -- does this mean that once 
> you have written your Parrot script and edited the generated source, 
> your GUI is essentially locked?

No.

> The problem as I see it is that the descriptive language is not the 
> primary tool: once you've generated your GUI source code (eg., in 
> Python), you're separated from the original script. There's no way to 
> re-generate the GUI framework and preserving your logic -- because there 
> is no separation here between presentation and code. You assume they're 
> be intermingled.

This is a good point.

One of my main motivations for wreiting Parrot is a dissatisfaction for
existing GUI building tools (at least the ones I've used).

This problem occurs with lots of GUI builders. One solution is to
put comments in the output of the GUI tool like this:

   //***** user code starts here
   //***** user code ends here

And then the GUI builder can read in the code the user has written.
I don't like this solution, IMO it looks ugly, and Parrot will use
a different solution: source code output by Parrot will go in a 
different file to source code written by the programmer. This is 
similar to tools like yacc -- the programmer doesn't directly alter the
file produced by this utility.

In Parrot, this will work by subclassing. Consider this except form
a .par file:


window @MyWindow {
   ...lots of other stuff here...
   button @start "Start"
   button @stop "Stop"
}


A Parrot backend will typically produce source code defining a class
MyWindow, in a particular programming language and GUI API, so a 
Python/Tkinter backend might produce something like this:


class MyWindow(DefaultWindow):
   ...lots of other code here...
   def start_pressed(self):
      # [Start] button pressed; this should be overridden 
      pass
   def stop_pressed(self):
      # [Stop] button pressed; this should be overridden 
      pass


Then the programmer writes their own file, which contains a subclass
of MyWindow, which contains method implementationd for start_pressed()
and stop_pressed().


> The issue is not just about good design, but also about practicalness: 
> If you run Parrot, its backend will (I conjecture) simply overwrite the 
> Python script,

Yes. But the programmer shouldn't directly be altering Parrot's
output script anyway.

> A more interesting approach would be to create a Python framework that 
> loaded such Parrot scripts and instantiated necessary objects at 
> runtime.

I intend to write a Parrot backend that outputs a file in Glade XML
format. I believe there already exist libraries that will do what
you suggest with Glade XML, so Parrot will be able to do it this
way. Parrot is intended to have multiple backends, so both approaches
can be used.

I am also intending to write a Glade XML frontend for Parrot, so
for example, one could use Glade to build the GUI, import the
Glade file into Parrot, and create Python/Tkinter code from this.

> For a good model on presentation/code separation, examine the Delphi VCL 
> framework. Your application is pure code; the rest is stored separately 
> in plaintext asset files, in a proprietary syntax not unlike your Parrot 
> language:
> 
> object MyForm: TMyForm
>   Left = 191
>   Top = 103
>   Width = 333
>   Height = 131
>   Caption = 'Form1'
>   Color = clBtnFace
>   Font.Charset = DEFAULT_CHARSET
>   Font.Color = clWindowText
>   Font.Height = -11
>   Font.Name = 'MS Sans Serif'
>   Font.Style = []
>   OldCreateOrder = False
>   PixelsPerInch = 96
>   TextHeight = 13
>   object OkButton: TButton
>     Left = 32
>     Top = 48
>     Width = 75
>     Height = 23
>     Caption = 'OK'
>     TabOrder = 0
>   end
>   object MyEdit: TEdit
>     Left = 32
>     Top = 16
>     Width = 209
>     Height = 21
>     TabOrder = 1
>     Text = 'This is an edit field.'
>   end
> end
> 
> (It's worth pointing out that, although you can edit these scripts, as a 
> rule you don't create these scripts yourself. They're generated by the 
> visual GUI builder. Most people don't even see these scripts.)

Glade XML format is similar (but is XML obviously); I will be looking 
into its suitability as a universal backend format for Parrot.
 
> The Delphi IDE itself has a neat two-way synchronization mechanism that 
> allows you to edit the code and still remain 100% in sync with the GUI. 
> For example, let's say you bind some code -- an "OnClick" event handler 
> -- to a button. You can edit the button, move it around, or edit the 
> event handler's code, move _that_ around -- but the link between the 
> source code and the GUI description is never severed.

Parrot will have thre same property. In Parrot .par format, ``@'' 
does this:

   button @myButton "Some Text"

means that the button will be known internally to the code as 
``myButton''; this is irrespective of where it appears on the 
screen.
 
> Please don't construe my comments as being anything but constructive 
> criticism. 

Youtr comments are welcome.

> I want great GUI tools for Python, too.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From earlybird at mop.no  Wed Aug 18 07:23:25 1999
From: earlybird at mop.no (Alexander Staubo)
Date: Wed, 18 Aug 1999 13:23:25 +0200
Subject: 16-bit Python -- does it exist?
References: <MPG.122214458f7ab0d2989685@news.mop.no> <37B81212.733CDD9B@interet.com>
Message-ID: <MPG.1224b6155e949206989686@news.mop.no>

In article <37B81212.733CDD9B at interet.com>, jim at interet.com says...
> Alexander Staubo wrote:
> > 
> > Subject says all. A friend (really!) would like to run Python on an old
> > 80286. Possible?
> 
> Python for 32-bit DOS and for Windows 3.1 are on the
> download page www.python.org/download.

Read the subject again. 32-bit Python won't run on a 286.

-- 
Alexander Staubo             http://www.mop.no/~alex/
"QED?" said Russell.
"It's Latin," said Morgan. "It means, 'So there you bastard'."
--Robert Rankin, _Nostramadus Ate My Hamster_



From arcege at shore.net  Wed Aug 18 08:02:13 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 18 Aug 1999 12:02:13 GMT
Subject: Controlling Python
References: <7p93ge$2bt$1@nnrp1.deja.com> <37B82DDA.2DE0E0A6@lemburg.com> <7pdr1h$gto$1@nnrp1.deja.com>
Message-ID: <9hxu3.498$mV.174274@news.shore.net>

nriedlin at my-deja.com wrote:

:> You can use a callback which you can register with the
:> sys.settrace() function. A less fine-grained hook is available
:> through sys.setprofile(). See the docs for more infos.
:>

: You are correct, but isn't it true, that the callback is only called
: every function call? I would like to interfere every line (and so being
: able to trace my scripts). Please correct me if I'm wrong.

: TIA

The settrace callback is envoked under four events: call, line, return,
and exception.  Below is a subsystem I wrote to help debugging.  Note that
there is a known "bug" in that call events do not return the arguments as
documented.

For more info, read:
  Python 1.5.2 Library Reference Manual 9.2 "How It Works"
  http://www.python.org/doc/current/lib/node201.html

  -Arcege

#!/usr/local/bin/python

import sys, traceback
import linecache

class Tracer:
  output = sys.stderr
  linecache = {}
  indent_value = 0

  def __init__(self, file=None):
    if file is None:
      pass
    elif type(file) is type(''):
      self.output = open(file, 'w')
    else:
      self.output = file

  def write(self, *args):
    apply(self.output.write, args)
    self.output.flush()

  def load_file(self, filename):
    if not self.linecache.has_key(filename):
      try:
        self.linecache[filename] = open(filename).readlines()
      except:
        self.linecache[filename] = []
  def lineno(self, filename, line):
    return linecache.getline(filename, line)
    self.load_file(filename)
    if line < len(self.linecache[filename]):
      return self.linecache[filename][line]
    return None

  def indent(self):
    self.indent_value = self.indent_value + 1
  def outdent(self):
    self.indent_value = self.indent_value - 1
  def output_indent(self):
    self.write('  ' * self.indent_value)

  def __call__(self, frame, event, arg):
    method_name = 'type_%s' % event
    getattr(self, method_name)(frame, arg)
    return self

  def type_call(self, frame, arg):
    self.output_indent()
    self.indent()
    self.write('Function call: %s' % frame.f_code.co_name)
    self.write('(')
    l = arg and len(arg) or 0
    for i in range(l):
      self.write(repr(arg))
      if i < l-1:
        self.write(', ')
    self.write(')')
    self.write('\n')
    return self

  def type_line(self, frame, arg):
    code = frame.f_code
    file = code.co_filename
    line = frame.f_lineno
    self.output_indent()
    self.write('Line number %d in %s (%s)\n  ->%s' %
                 (line, file, code.co_name, self.lineno(file, line)))
    return self

  def type_return(self, frame, arg):
    self.outdent()
    self.output_indent()
    self.write('Function %s returning %s\n' %
                 (frame.f_code.co_name, repr(arg)))

  def type_exception(self, frame, arg):
    code = frame.f_code
    funcname = code.co_name
    exc, value, traceback = arg
    t = type(exc)
    if t is type(''):
      name = exc
    elif t is type(self):
      name = exc.__class__.__name__
    elif t is type(self.__class__):
      name = exc.__name__
    else:
      name = repr(exc)
    self.output_indent()
    self.write('Exception %s raised in %s\n' % (name, funcname))
    self.outdent()
    return self.inside_exception

  def inside_exception(self, frame, event, arg):
    self.write('Exception: %s\n' % event)
    return self.inside_exception

def settrace(tracer):
  # Start debugging from here
  try:
    raise SystemError
  except:
    if hasattr(sys, 'exc_info'):
      frame = sys.exc_info()[2].tb_frame.f_back
    else:
      frame = sys.exc_traceback.tb_frame.f_back
  while frame:
    frame.f_trace = tracer
    frame = frame.f_back
  sys.settrace(tracer)

def unsettrace():
  settrace(None)

class _Num:
  def __init__(self, value=None):
    if value is None:
      self.value = self.start
    else:
      self.value = value
  def __str__(self):
    return str(self.value)

class _Sigma(_Num):
  start = 0
  def __add__(self, other):
    self.value = self.value + other
    return self

class _Pi(_Num):
  start = long(1)
  def __mul__(self, other):
    self.value = self.value * other
    return self

def _test(count):
  sum = _Sigma(); aggr = _Pi()
  for i in xrange(count):
    j = i + 1
    sum = sum + j
    aggr = aggr * j
  print 'Sigma(%d)' % count, sum
  print 'Pi(%d)' % count, aggr

if __name__ == '__main__':
  tracer = Tracer()
  settrace(tracer)
  _test(100)




From philh at vision25.demon.co.uk  Sat Aug 21 18:05:07 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sat, 21 Aug 99 22:05:07 GMT
Subject: can some one help a newbie?
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <7pkke6$lut@dfw-ixnews6.ix.netcom.com> <935227166snz@vision25.demon.co.uk> <7pmv4j$gtt@dfw-ixnews14.ix.netcom.com>
Message-ID: <935273107snz@vision25.demon.co.uk>

In article <7pmv4j$gtt at dfw-ixnews14.ix.netcom.com>
           aahz at netcom.com "Aahz Maruch" writes:
> In article <935227166snz at vision25.demon.co.uk>,
> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> >In article <7pkke6$lut at dfw-ixnews6.ix.netcom.com>
> >           aahz at netcom.com "Aahz Maruch" writes:
> >>
> >> Overall, if your background is as a sysadmin using shell scripts, grep,
> >> awk, sed, and so on, you'll find Perl a bit easier.
> >
> >I've used all these, and find Python easier than Perl. Perl would
> >probably be easier than Python for 1-liners, if I could remember
> >the syntax.
> 
> Note that I'm specifically assuming a sysadmin with little actual
> programming experience.

The idea that a company would employ someone, who can't really code,
as a sysadmin is IMO frightening. Unless they don't care whether
their computer system works, of course.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From kernr at mail.ncifcrf.gov  Wed Aug 11 00:31:24 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Wed, 11 Aug 1999 04:31:24 GMT
Subject: [OT] Out-of-Office Autoresponders and Mailing Lists Don't Mix
Message-ID: <37b0fa0b.13118215@news.erols.com>

Sorry for the offtopic post, but I've noticed three separate instances
of people on three different Python-related mailing lists using
"out-of-office" e-mail autoresponders while still receiving messages
from the list.

All of these lists are now run by Mailman, which has a convenient
option for disabling delivery for a time.  Please use it if you also
use an e-mail autoresponder.

sighing-because-the-people-that-should-read-this-are-on-vacation-ly
y'rs

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From fdrake at acm.org  Fri Aug 20 11:08:00 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 20 Aug 1999 15:08:00 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
In-Reply-To: <37bcb30b@cs.colorado.edu>
References: <37B47506.14EBD269@msfc.nasa.gov>
	<c75yafced1p.fsf@pc142.cosc.canterbury.ac.nz>
	<37bb6923@cs.colorado.edu>
	<c75g11f1doq.fsf@pc142.cosc.canterbury.ac.nz>
	<37bcb30b@cs.colorado.edu>
Message-ID: <14269.28496.156064.652501@weyr.cnri.reston.va.us>

In comp.lang.python, 
    Timothy R Evans <tre17 at pc142.cosc.canterbury.ac.nz> writes:
 >def wibble():
 >    print x
 >    x = 6 # this fails with NameError
 >
 >Seems like a good behaviour as it stops you doing dumb things.

Tom Christiansen writes:
 > Perhaps it could give a better error message, pretty please?  Sometimes
 > driving Python feels  like driving Ken Thompson's car.  It just keeps saying
 > "?" and you're expected to know what that means.  :-(

  Current behavior (1.5.2 as released) raises UnboundLocalError (a
subclass of NameError), which is much more informative and is usually
exactly what is needed to point out to the programmer where the coding 
error is.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From bowman at montana.com  Sat Aug 21 00:53:14 1999
From: bowman at montana.com (bowman)
Date: Fri, 20 Aug 1999 22:53:14 -0600
Subject: Book -- or website -- about Tk & tkinter
References: <935172552snz@vision25.demon.co.uk>
Message-ID: <37BE30BA.E45BB293@montana.com>

Phil Hunt wrote:
> 
> Can anyone recommend a good book or website that explains how Tk does
> its geometry management (the grid, pack etc managers)?

Do you mean the low level nuts and bolts, or just how to use the
managers effectively? Brent Welch's _Practical Programming in Tcl and
Tk_ is the best book I've seen at the local bookstore, which isn't
saying much. At least 300 of the 400 pages deal with Tk, and he does
have some good hints and examples on building up a GUI with frames.

Many of the concepts are similar to Java, so any of the Java GUI
tutorials might be useful, or, since Tk was a veneer on the X toolkit,
some of the X docs. The basic concept is the old divide and conquer
technique. You can stuff a couple of widgets into a frame and get them
to behave. Then you put that frame into another frame, and deal with
that layer of siblings, and so forth. 

As far as tkinter, read the life preserver, Fredrik Lundh's in progress
work, and the code in the Demo and Tools packages. Mostly, I'm afraid,
read between the lines, and experiment.  
-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From nhead at houbits.com  Wed Aug  4 04:03:25 1999
From: nhead at houbits.com (Nigel Head)
Date: Wed, 04 Aug 1999 10:03:25 +0200
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk>
Message-ID: <37A7F3CC.583F2C7F@houbits.com>

Ian Clarke wrote:

> I am presently working on a CGI based system in Python, and the
> experience gave me an idea for a simple Python program, I would
> appreciate any comments (such as "don't bother, it's been done" or
> "sure, I have loads of spare time, I will do it and release it under the
> GPL").

I'm not going to say either of those! I have done it, although I'm not
terribly happy with the results so far and it's on the back burner for
the moment .... I need to figure out if it's still useful in a Zope
world - which means I have to learn and understand Zope!

> It is to create a Python web server which automatically parses embedded
> Python in HTML (there is a Perl extension that does this too).  So, for
> example, the following HTML code:

I embed the python fragments in HTML comment blocks for the sake of
my HTML editor tools which would otherwise feel very sad ...

I've also set it up (at least in one version) so that the python can
generate
http headers at the start as well .... or they can be embedded in the
templates.

> Internally, operation of the system would be simple.  The HTML page with
> embedded python would first be converted to a python file (by having a
> print statement output the HTML between the embedded python), this could
> then be compiled for efficiency as when normal files are executed.  The
> web server would then parse these files when requested (using an
> appropriately neat method to allow access to CGI variables and Cookies
> and the like from within the embedded HTML).  This would lead to a speed
> improvement in that the Python interpreter need not be launched every
> time.  The efficiency could be further improved by caching the compiled
> python files in memory.

Yup! All there (minus "appropriately neat": more "rather nasty" for now I'm
afraid)

> Cool facilities could also be implemented which sit ontop of the base
> CGI and Cookie layers to emulate a "per-user" memory.  This would allow
> a dictionary to be stored only to be retrieved the next time this user
> accesses the site (this would be achieved using cookies and storing
> information locally with a cookie-style self destruct).  Similarly, a
> "per-session" memory could be implemented, making it much easier to
> transfer information (such as a username and password) from one page to
> the next.

Hhmm! Didn't tackle this sort of thing yet !

Also I haven't yet dealt with putting the python into rexec for execution-
but this is only needed if a web master wants to give their clients access
to
these facilities as opposed to doing things themselves.

Things I'm not happy with:

* indenting of the python bits versus the HTML bits doesn't mesh
   very well

* the base server that I use is a threaded version of SimpleHTTP and I seem
   to have occasional problems with the thing hanging ... don't know why
   and haven't time to look into it just now

* I haven't figured how many security holes this might open ...

* there aren't any database "adapters" yet -- I just use background
    threads with an in memory database shelved to flat files when
    written to. Locking between threads is there though to allow
    this to be done safely ... Gadfly would not be a difficult step!

* basically it isn't anywhere near finished

* there are no good examples in the package yet

Having said that I have a couple of small sites where the thing is running
happily
with low load levels. Runs on NT/95 and BSD OS3 as of today. Used to run
on an very old RedHat linux but hasn't been tested there in a while!

I'll upload the thing to my starship account when I get home tonight just
in case anyone is interested!! I'll get back tomorrow with a URL.

Nigel.

nhead at houbits.com, nhead at esoc.esa.de




From paul at prescod.net  Thu Aug 12 11:11:40 1999
From: paul at prescod.net (Paul Prescod)
Date: Thu, 12 Aug 1999 15:11:40 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com>
		<37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk>
Message-ID: <37B2E42C.6A377176@prescod.net>

Phil Hunt wrote:
> 
> That is a persuasive argument. One counterargument would be that
> any application that uses that approach would need a built-in
> XML parser and Parrot run-time library, thus creating code bloat;

One thing I've noticed is that sometimes you write these code generators
and then find that you still need a runtime library and it isn't much
smaller than the one you would have done directly. 

Plus you have the cost of the duplicated redundant code created by the
code generator. So then you look at that duplicated code and you say: "I
should move as much as possible into the runtime library." So the
runtime library gets larger and eventually you say: "The 'generated
code' is really just a declarative specification that depends on a ton
of runtime code. So then you say: "I'm just converting one declarative
specification to another. That doesn't seem right."

As far as the size of the parser goes, there are two ways to minimize
that. You can either use a syntax that the language already has a parser
for (increasingly the case with XML) or you can use a syntax that is
really trivial to parse (Lisp S-exprs?).

 Paul Prescod



From para at o3.net  Sat Aug  7 04:31:25 1999
From: para at o3.net (Para Hsu)
Date: Sat, 7 Aug 1999 16:31:25 +0800
Subject: Python GUI
Message-ID: <01bee0af$3cdbf7e0$LocalHost@para.cis.scu.edu.tw>

Spam detection software, running on the system "albatross.python.org", has
identified this incoming email as possible spam.  The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email.  If you have any questions, see
the administrator of that system for details.

Content preview:  -----???l?l??----- ?H????: Kevin Dahlhausen <morse at harborcom.net>
   ?s?D?s??: comp.lang.python ??????: python-list at python.org <python-list at python.org>
   ????: 1999?~8??6?? PM 08:19 ?D??: Re: Python GUI >There is a TK/TCL builder
   called 'SpecTCL.' that has a Python module called >"SpecPython." I don't
  have the URL though. It works very well, if you like [...] 

Content analysis details:   (6.9 points, 5.0 required)

 pts rule name              description
---- ---------------------- --------------------------------------------------
 2.0 FH_DATE_IS_19XX        The date is not 19xx.
 0.7 XMAILER_MIMEOLE_OL_CAC8F XMAILER_MIMEOLE_OL_CAC8F
 4.2 FORGED_MUA_OUTLOOK     Forged mail pretending to be from MS Outlook


-------------- next part --------------
An embedded message was scrubbed...
From: "Para Hsu" <para at o3.net>
Subject: Re: Python GUI
Date: Sat, 7 Aug 1999 16:31:25 +0800
Size: 1418
URL: <http://mail.python.org/pipermail/python-list/attachments/19990807/635b64fd/attachment.mht>

From tchrist at mox.perl.com  Wed Aug 18 23:22:16 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 18 Aug 1999 20:22:16 -0700
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <000501bee6cb$a31710e0$832d2399@tim> <37B675A9.53C2BBE0@getaway.net> <slrn7rdlg1.en.amk@207-172-57-199.s199.tnt2.ann.va.dialup.rcn.com>
Message-ID: <37bb6a58@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, akuchlin at mems-exchange.org writes:
:... but if you coded "for i
:in biglist: resultlist += thingy", you'd do a lot of unnecessary list
:copying.  

Only if you implemented it that way, which wouldn't appear to be
particularly clever. :-)

--tom
-- 
    "Just because you're screwed *up* doesn't mean you're screwed." --Larry Wall



From tim_one at email.msn.com  Tue Aug 17 00:32:46 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 00:32:46 -0400
Subject: Bug in struct module?
In-Reply-To: <37B8759C.D0DF1B55@bigfoot.com>
Message-ID: <000001bee869$8e9cbf80$6e2d2399@tim>

[Hoon Yoon]
> >>> calcsize('hd')
> 16
> >>> calcsize('dh')
> 10
> That seems like a bug to me.

Assuming you're running on Win9x/NT, it's functioning as designed, although
the actual size of a double-followed-by-short struct under MSVC's default
rules is 16 instead of 10.  To get the actual total size of a struct under
MSVC's defaults, you need to append "0{x}" to the format, where {x} is the
typecode of the largest of the preceding typecodes:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
IDLE 0.5 -- press F1 for help
>>> import struct
>>> struct.calcsize("hd0d")
16
>>> struct.calcsize("dh0d")
16
>>>

If you'd rather get 10 for each result, use the "native byte order, standard
alignment" format code:

>>> struct.calcsize("=dh")
10
>>> struct.calcsize("=hd")
10
>>>

See the docs for details on that.  If you're not using native C structs, the
"standard alignment" rules are much easier to live with (they never
introduce hidden pad bytes, while native alignment may).

but-first-figure-out-how-the-data-is-layed-out-ly y'rs  - tim





From tim.jarman at lineone.net  Fri Aug 20 15:31:26 1999
From: tim.jarman at lineone.net (Tim Jarman)
Date: Fri, 20 Aug 1999 19:31:26 +0000
Subject: carpe linguam
References: <199908201554.LAA11430@python.org>
Message-ID: <37BDAD0A.FAFC980A@lineone.net>

Tom Christiansen <tchrist at mox.perl.com>wrote:

>> I studied Spanish, French, German, Italian, Portuguese, or Latin

Of course, having studied Latin, then Spanish, Italian and Portuguese are trivial... and partially explains the misquotation of Horace in the title of this post.

classically y'rs

Tim Jarman
"A smile, a song, and a compiler"




From eye at niksula.hut.fi  Fri Aug 20 14:09:55 1999
From: eye at niksula.hut.fi (Petri Mikael Kuittinen)
Date: 20 Aug 1999 21:09:55 +0300
Subject: How to sell Python to your boss?
References: <ka8yaf7d84t.fsf@delightful.cs.hut.fi> <wjywvur2qrn.fsf@bradh.ne.mediaone.net>
Message-ID: <ka8so5ez5l8.fsf@delightful.cs.hut.fi>

Brad Howes <bradh at mediaone.net> writes:

> Heck, I gave up and joined a company (moving ~2800 miles) that hired me
> *because* of my Python skills.

I have also received two work offers concerning Python programming. I
wasn't interested in the first one and the second was was such a short
project, that it is not worth to give my current work for it.

It is a kinda funny. When I started to learn Python at the end of last
year, many people laughed at me: "How can you have time and energy to
learn a language that nobody uses and we have never heard of? No
employer uses Python anywhere. Java is so great, the best language,
you better forget Python and start using it...".

It is true that a Java guru might have easier time to find a job (I
have done only a little bit of programming in Java), but there is
demand for Python programmers as well.


Petri

-- 
<(O)> Petri Kuittinen, also known as Eye, Dj Eye or Peku               <(O)>
<(O)> ADDRESS: Timpurinkuja 1 C 39, FIN-02600 Espoo, Finland           <(O)>
<(O)> EMAIL: eye at iki.fi WWW: http://www.iki.fi/~eye/ PHONE: 09-5472380 <(O)>
~KISS = Keep It Simple Stupid



From fredrik at pythonware.com  Fri Aug 27 03:29:08 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 27 Aug 1999 07:29:08 GMT
Subject: count method only for lists?
References: <37C57DAB.63C7@cs.bham.ac.uk>
Message-ID: <00ea01bef05d$da7d6bb0$f29b12c2@secret.pythonware.com>

Stuart Reynolds <S.I.Reynolds at cs.bham.ac.uk> wrote:
> Does anyone know why there the count method is only defined for mutable
> sequences (i.e. lists) and not strings and tuples?

"Good question", to quote the FAQ.

in the old days, the answer used to be 'the design rule is that
immutable objects don't have methods', but that's no longer
true.  strings are likely to grow methods in 1.6, for example. 
and complex numbers already have them...

FAQ 6.4 gives the true background:

    http://www.python.org/doc/FAQ.html#6.4

    Long ago, it seemed unnecessary to implement any of
    these functions in C, so a standard library module "string"
    written in Python was created that performs string
    related operations.

> It seems as though the Python Library Reference (2.1.5.2
> Mutable Sequence Types) only describes lists and not
> mutable sequence types.

well, strings and tuples are not mutable, so that's
reasonable.

</F>




From wware-nospam at world.std.com  Wed Aug 25 22:41:41 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Thu, 26 Aug 1999 02:41:41 GMT
Subject: C module leaking memory
Message-ID: <FH1wtI.C68@world.std.com>

I'm working on a fairly complicated C module and it appears
to be leaking memory. It runs a while, and then the interpreter
gives up, saying 'MemoryError'. Are there any good pointers for
debugging memory leaks? I've been pretty careful about watching
mallocs and frees, and Py_INCREFs and Py_DECREFs. Are there other
good places to look, or should I continue looking more closely
at those? Any wisdom would be much appreciated, thanks.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From harryt at gte.net  Thu Aug 26 10:58:09 1999
From: harryt at gte.net (Harry Tanovich)
Date: Thu, 26 Aug 1999 14:58:09 GMT
Subject: Off Topic: bsddb 1.85
Message-ID: <37C557E2.81B8A173@gte.net>

When doing len(my_bsddb_instance) on a 40MB bsddb file, my RAM filled up
before I got my answer back.  Does bsddb ever re-use it's cache or does
it just keep putting things into RAM?




From xgzplp at yahoo.com  Sat Aug 14 19:18:34 1999
From: xgzplp at yahoo.com (xgzplp at yahoo.com)
Date: 14 Aug 1999 15:18:34 -0800
Subject: XXX Fucking Pics
Message-ID: <37b5eb3a$3@news.vphos.net>

XXX PICS XXX PICS
http://www.nettaxi.com/fcitizens/adult168/dp/p3.htm




From pinard at iro.umontreal.ca  Sat Aug 28 08:30:08 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 28 Aug 1999 08:30:08 -0400
Subject: Hello, world!
Message-ID: <oqzozcf5pr.fsf@titan.progiciels-bpi.ca>

Hello, people.

Let me say hello!  I'm a recent Python convert (OK, OK: newbee! :-).
I've been reading the message for the last week.  Once put aside random
inelegance in the Perl-Python thread, I like the spirit of the gang so far.
Articles are usually quite informative, and there is kindness in the
attitudes.  I merely wanted to tell you.

						Keep happy, all.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From Alessandro.Bottoni at think3.com  Tue Aug  3 08:04:41 1999
From: Alessandro.Bottoni at think3.com (Alessandro Bottoni)
Date: Tue, 3 Aug 1999 12:04:41 GMT
Subject: Any Python interpreter for NOKIA 9110?
Message-ID: <6D8A17398E28D3119F860090274DD7DB0641BD@pces.cadlab.it>

Does anybody knows if there is a Python interpreter for the Nokia 9110
mobile phone?

(Nokia 9110 runs the Geoworks GEOS Operating System. 
See http://www.geoworks.com/devices/9110/)

Thanks

--------------------------------
Alessandro Bottoni
(Alessandro.Bottoni at Think3.com)
(abottoni at yahoo.com)
Web Programmer
Think3 inc.
(www.think3.com)




From infody864 at my-deja.com  Sun Aug  8 22:37:34 1999
From: infody864 at my-deja.com (infody864 at my-deja.com)
Date: Mon, 09 Aug 1999 02:37:34 GMT
Subject: NatLink
Message-ID: <7oletd$jdj$1@nnrp1.deja.com>

I would be interested in conversing with programmers for using Python
and NatLink.
If anyone is interested in helping each other, please contact me.

Kind regards,

Bob Dempsey
infody at earthlink.net


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From forcer at mindless.com  Tue Aug 24 05:44:19 1999
From: forcer at mindless.com (forcer)
Date: 24 Aug 1999 11:44:19 +0200
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan> <slrn7s408d.gvn.wtanksle@dolphin.openprojects.net>
Message-ID: <hhvha5tswc.fsf@forcix.roof.lan>

wtanksle at dolphin.openprojects.net (William Tanksley) writes:

> forcer at mindless.com (forcer) writes:
> >   def f(arg): pass    # a function that does nothing (yet)
> 
> >one could write
> 
> >	if foo:
> >	bar
> 
> >Syntactically this does not introduce any ambiguity. The only
> >problem i see is that the "pass" version is more readable.
> 
> Precisely.  You're suggesting a change with the sole purpose of reducing
> redundancy.  The problem is that humans _need_ redundancy.  Not only does
> redundancy help us to read the code, it also helps the computer to find
> our errors.

The important point here is that the necessity of ``pass'' gets
into your way in some places. I agree that some redunancy is nice
if it helps to read the code, but it shouldn't get into your way.

You didn't quote the example where i pointed out the real
problem with pass:

if foo:
   # bar
   # baz
qux

E.g. if you're developing code and just commented out a (not yet
worked out branch) to work on a different part of the program.
Requireing pass here is unecessary.

I didn't ask for pass to vanish completely, i merely asked for it
to optional instead of mandatory.

> Another example of redundancy in Python -- and one with just as much
> chance of being changed -- is the ':' symbol.  Take a look at the following:

Though the colon really improves readability while not getting in
your way :)

	-forcer

-- 
((email . "forcer at mindless.com")       (www . "http://forcix.cx/")
 (irc   . "forcer@#StarWars (IRCnet)") (gpg . "/other/forcer.gpg"))



From paul at prescod.net  Thu Aug 26 13:33:52 1999
From: paul at prescod.net (Paul Prescod)
Date: Thu, 26 Aug 1999 17:33:52 GMT
Subject: Library philosophy
Message-ID: <37C57A80.D5C62704@prescod.net>

I am curious about the opinions of Python community members on the
following issue:

I believe that Python's libraries are in many places too thin of a layer
on top of the C library and are thus not very Pythonish. Evidence of
this includes indecipherable function names like strtfmt and APIs that
return tuples instead of objects and take integers as "flags" instead of
strings or objects.

Would it it make sense to start wrapping the low-level modules with
high-level, easier to use modules that would be "the default" in Python
2?

 Paul Prescod



From tim_one at email.msn.com  Sat Aug 21 22:47:02 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 21 Aug 1999 22:47:02 -0400
Subject: ANNOUNCE normalDate 0.7
In-Reply-To: <37BEE360.2D96EB9D@rubic.com>
Message-ID: <000301beec48$9d0fa040$cb2d2399@tim>

[Jeff Bauer]
> ...
>   http://starship.python.net/crew/jbauer/normalDate/
> ...

Jeff, that URL didn't work for me.  This one did:

    http://starship.python.net/crew/jbauer/normaldate/

I think the server is case-sensitive.

> ...
> The Python distribution also includes a date class, which would
> have saved me time, had I bothered to check.
>
>  Demo/classes/Dates.py

As the author of Dates.py, let me recommend Jeff's NormalDate.py.  Dates.py
was written as a test drive of then-new operator overloading facilities, not
as a real date class (that's why it's under the Demo tree).  Jeff's has more
features and is more practical, but is also very small and isn't bogged down
with esoterica either.  OTOH, his does have a Y10K problem, while mine works
until the year gets so big it won't fit in an unbounded Python long, and
even then you'll get a butt-saving MemoryError <wink>.

use-dates.py-if-you're-immortal-ly y'rs  - tim





From bwarsaw at cnri.reston.va.us  Tue Aug  3 22:45:17 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 3 Aug 1999 22:45:17 -0400 (EDT)
Subject: Q: names with underscores and style
References: <37A79947.187EAB62@arlut.utexas.edu>
Message-ID: <14247.43325.179807.148034@anthem.cnri.reston.va.us>

>>>>> "TAB" == Thomas A Bryan <tbryan at arlut.utexas.edu> writes:

    TAB> Basically, I'm trying to figure out when to use a leading 
    TAB> underscore for a method/member name.  Is there some rule of
    TAB> thumb for underscores in Python names?

Attributes with DLUNTU (double leading underscore, no trailing
underscores) are treated as private attributes, and Python munges
their names.  E.g. self.__privatemeth()  Use these when you definitely 
don't want subclasses to access or override them (Python doesn't
actually /enforce/ any access control -- remember we're all consenting 
adults :).

My own personal convention is to use SLUNTU (single leading ...) when
I want attributes that are part of the `protected' interface of a
class.  E.g. self._override_this()

DLUDTU (double leading ... double trailing), e.g. self.__init__() are
reserved for Python's use.

-Barry



From ivanlan at callware.com  Thu Aug 26 12:37:22 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Thu, 26 Aug 1999 16:37:22 GMT
Subject: [Monterey-speakers] Re: [monterey] [Monterey-speakers] Thanks!
References: <Pine.WNT.4.05.9908252026040.130-100000@david.ski.org> <199908261600.MAA01158@eric.cnri.reston.va.us> <37C566CE.E3F19D3B@callware.com>
Message-ID: <37C56D42.609DBA38@callware.com>

Hi All--

I'm cross-posting this to the regular Python list too.  To clue them in,
here is some information on the audio tapes available for the O'Reilly
Open Source convention and the Python track:


http://www.vwtapes.com/cgi-local/vwcart.pl?page=/oreilly/99/keynotes.html&cart_id=4733886.202871

lists the audio tape for Guy's speech.

And the page for the Python tapes is:


http://www.vwtapes.com/cgi-local/vwcart.pl?page=/oreilly/99/python.html&cart_id=4733886.202871

The main page is

	http://www.vwtapes.com

<mike-the-man-behind-the-curtain>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From da at ski.org  Sat Aug 14 17:19:48 1999
From: da at ski.org (David Ascher)
Date: Sat, 14 Aug 1999 14:19:48 -0700 (Pacific Daylight Time)
Subject: Regarding the magical "or" operator
In-Reply-To: <37b5d712@cs.colorado.edu>
Message-ID: <Pine.WNT.4.05.9908141412520.157-100000@david.ski.org>

Hi Tom!

On 14 Aug 1999, Tom Christiansen wrote:

> I was wondering how often most python programmers make use of that
> remarkably nifty property of the logical operators, in which they
> produce not merely 0 or 1, but the value is last looked at when it
> decided whether the expression was true or false.

I'm not sure about most python programmers.  It's a feature of the
language that I always teach in my courses, and I tend to get blank looks
at first.  I find that once I motivate the or behavior with an example
like:

    title = windowname or 'Untitled'

a few folks think it's neat (and I guess from that that they plan on using
it, or at least trying it).  I use it periodically, although not very
often.

Not that you asked, but I find the 'and' behavior much harder to motivate
and grok, and I avoid it personally.

> My question is this: have you ever had an urge for it (or a variant)
> to care about the difference between None and any other false value?
[...]
> The idea would be a version of "or" that would return the first non-None
> value.  For example, supposing this were called "nor" (silly name,
> but this is *not* a proposal):
> 
> 	print f() nor h()
>     []
> 
> 	print g() nor h()
>     ['Some']
> 
> The argument is that this would be used not with prints, but with
> assignments, as in
> 
>     a = f() nor g()
> 
> The reason I ask is not because I actually want such a thing.  Gosh, no.
> I'm rather uncomfortable with the idea.  I'm just trying to gather some
> understanding and perspective for use with the folks who *would* like
> such an oddity. :-)

FWIW, it doesn't grab me.  I'd read 'a nor b' as 'not (a or b)', and I
know too much about truth and falsehood in Python to expect None to be
treated differently.

(It's a nice word, though.  One that, alas, Dr. Seuss didn't use when he
should have in One Fish, Two Fish, Red Fish, Blue Fish.)

--david ascher






From skip at mojam.com  Tue Aug 24 13:07:21 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 24 Aug 1999 12:07:21 -0500 (CDT)
Subject: Extending Python: exception handling
In-Reply-To: <3.0.6.32.19990824122931.0099e860@mail.triode.net.au>
References: <37bdcb30.796706964@news.triode.net.au>
	<14271.18469.757616.424717@dolphin.mojam.com>
	<3.0.6.32.19990824122931.0099e860@mail.triode.net.au>
Message-ID: <14274.52504.789825.306617@dolphin.mojam.com>

I wrote:

    >It makes a reasonable substitute for the lack of labelled break
    >statements on certain occasions.
    >
    >    class BreakLabel: pass
    >    try:
    >	    for i in ...:
    >	       for j in ...:
    >		   exit_cond = do_some_work()
    >	 	   if exit_cond: raise BreakLabel
    >    except BreakLabel:
    >	    pass

to which John replied:

    Two questions:

	    1) would you always use a user defined exception?
	    2) What if you had labelled breaks?

    Reason: suppose I implement labelled statements in Viper,
    and suppose I assume that system exceptions (Attribute
    errors, etc) are program faults -- would all your code work?

1. Since I'm not catching errors and I might want to have multiple "break
   labels", yes I would always use a user-defined exception.

2. If I had labelled breaks I'd use 'em.  I'm obviously overloading
   exceptions because I don't have break labels.  I find the above a bit
   clearer than keeping track of where I'm going with extra state variables
   however.

Yes, my code would work.  I try to use the system's builtin exceptions as
they were intended - to catch problems.

Later on, John wrote:

	    try:
		    y = sqrt(x)
	    except:
		    y = x.sqrt()

    Here, the programmer is assuming that if the builtin sqrt function
    defined for floats fails, the object might be a class emulating numbers,
    and try to call a sqrt method.  But the Viper compiler might _know_ that
    x is a class instance and that the builtin function 'sqrt' only applies
    to floats, and report an error.

    Even worse, it might _deduce_ that x must be a float, and then get
    confused seeing x.sqrt(), which might require that x is an instance
    (depending on whether sqrt is a method of floats or not).

Then you're programming in a language other than Python.  If you can
reasonably determine that x is an instance and not a float, I'd expect the
compiler to generate the equivalent of simply

    y = x.sqrt()

if you know that executing the try block will always fail.

If you can deduce it's a float I'd expect it to generate the equivalent of
the above try/except statement, since the programmer didn't qualify the
except clause.  You could get either an AttributeError or an OverflowError
from sqrt.  If it had been:

    try:
	y = sqrt(x)
    except AttributeError:
	y = x.sqrt()

then you could generate just

    y = sqrt(x)

when x is a float, because an AttributeError can't occur.  (Python currently
raises an AttributeError if an instance doesn't have a __float__ method.)

Further down:

    The basic problem is that in the presence of dynamic typing and
    exception handling, type inference is hard, especially because handlers
    for exceptions can be lexically isolated from the point the exception is
    raised: with a _complete_ flow analysis, the type inference system
    wouldn't work in all cases. [and maybe not even then]

Yup.  You aren't going to be able to do much other than perhaps spit out a
warning to the programmer that no optimization is being performed because
the except clause is unqualified.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From gmcm at hypernet.com  Sun Aug  8 11:17:10 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 8 Aug 1999 10:17:10 -0500
Subject: Problem with modules refering to each other
In-Reply-To: <934075710snz@vision25.demon.co.uk>
Message-ID: <1278000860-36223465@hypernet.com>

Phil Hunt writes:

> I have two python modules, both of which need to refer to classes
> defined in the other one. How do I do this? 
> 
> I am currently putting at the start of each module a line of the
> form:
> 
>    from theOtherModule import *
> 
> However, this is not working. My modules -- greatly simplified --
> look like this:
> 
>    #bak.py
>    from hhh import *
>    ti = AnHhhClass()
>    print ti
> 
> and:
> 
>    #hhh.py
>    from bak import *
>    class AnHhhClass: pass
> 
> When I run ``python bak.py'', I get the error message 
> ``NameError: AnHhhClass''.

Plumbing rule #1: bak.py runs, Python hits the line with the "import" 
statement, moves over to hhh.py. In hhh.py it hits the "import" line. 
To protect itself against infinite recursion, it looks to see if 
bak.py is being imported. In your case it's not, because what you 
think of as "bak" is actually "__main__". So it starts importing bak 
as a module. Hits the import line. Looks up hhh.py. This time it 
detects recursion, so it just goes on, on the theory that whoever is 
importing hhh.py will eventually complete the process. Which is true. 

Plumbing rule #2: from module import ... doesn't do what you want. It 
imports module, but not into your namespace. It takes what it finds 
in the imported namespace and creates vars of the same name in your 
namespace.

So when bak.py is imported, nothing has been bound into it's
namespace from hhh, because hhh's namespace is empty at that point.
Since the following line expects something from hhh to be bound into
bak.py's namespace, it fails.

So how do you fix this? 

First, you never, ever use "from module import ..." unless (1) 
the module has been specifically designed to allow this (see 
what Tkinter or threading does to allow this) and (2) there are no 
recursive imports going on.

Second, if you're doing recursive imports, you don't make any 
references to items in the other module at the top level. They won't 
be resolvable until the top level has been completely scanned.

Third, you remember that the script you run is not imported under 
it's own name. It's in the import list, but as "__main__". Forgetting 
this yields 2 copies of your module, and strange errors where vars 
you set to new values don't change.

So bak.py should say:

import hhh

def doit()
  it = hhh.AnHhhClass()
  print it

And you should run a script that says:

import bak

bak.doit()

- Gordon



From ionescu at pik-potsdam.de  Fri Aug 27 07:25:56 1999
From: ionescu at pik-potsdam.de (Cezar Ionescu)
Date: Fri, 27 Aug 1999 13:25:56 +0200
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com> <001f01bef078$25fcf870$f29b12c2@secret.pythonware.com>
Message-ID: <37C675C4.F12C011E@pik-potsdam.de>

Fredrik Lundh wrote:
> 
> <guy_oliver at yahoo.com> wrote:
> > Why do we call python a scripting language?
> 
> who's calling it that? ;-)
> 

>From the backcover to Learning Python by Mark Lutz and David Ascher:

Learning Python is the perfect introduction to this increasingly popular
scripting language.

Cheers,
Cezar.



From mspal at sangria.harvard.edu  Fri Aug  6 06:28:20 1999
From: mspal at sangria.harvard.edu (Michael Spalinski)
Date: 06 Aug 1999 10:28:20 +0000
Subject: ANN: PIL 1.0 final
References: <008e01bedfec$7ffe7110$f29b12c2@secret.pythonware.com> <37aadb42.108851580@news.u-psud.fr>
Message-ID: <m37ln9i4or.fsf@schwinger.harvard.edu>

I compiled PIL 1.0 on Red Hat Linux 5.2, but then I get:

[schwinger:/usr/src/python/Python-1.5.2/Extensions/Imaging-1.0]# python
Python 1.5.2 (#5, Apr 14 1999, 19:21:58)  [GCC 2.7.2.3] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import _imaging
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: /usr/lib/libtk8.0.so: undefined symbol: XFreePixmap
>>>   

Is my tk version too old? I have tk-8.0.3-20.

M.





From skip at mojam.com  Thu Aug 19 14:44:49 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 19 Aug 1999 13:44:49 -0500 (CDT)
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc4daf@cs.colorado.edu>
References: <37bc1339@cs.colorado.edu>
	<8E27B3FC4duncanrcpcouk@news.rmplc.co.uk>
	<37bc4daf@cs.colorado.edu>
Message-ID: <14268.20536.389594.471726@dolphin.mojam.com>

    Tom> BTW, the @colors assignment is more legibly written using the qw//
    Tom> syntactic sugar:

    Tom>     @colors = qw(red blue green yellow orange purple);

In Python an occasional idiom you see is to use string.split to save typing
so many quotes:

    colors = string.split("red blue green yellow orange purple")

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From jstevens at basho.fc.hp.com  Thu Aug  5 17:27:18 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: Thu, 05 Aug 1999 15:27:18 -0600
Subject: RMI and Python
References: <Pine.GSO.4.10.9908011131490.25758-100000@xmission.xmission.com>
Message-ID: <37AA01B6.A54E2B23@basho.fc.hp.com>

Tim Egbert wrote:
> 
> Java is able to do remote method invocation (RMI) which is a neat concept.
> It is the basis for Jini, another neat idea.  Does Python have anything
> like RMI?
> 
> Tim Egbert

I've a simple Distributed Objects package you could try, if you
like.

Email me if you'd like to get a copy.

John S.



From samschul at pacbell.net  Sat Aug 28 14:11:09 1999
From: samschul at pacbell.net (Sam Schulenburg)
Date: Sat, 28 Aug 1999 11:11:09 -0700
Subject: IDLE: What to use instead of sys.argv[0] ?
References: <MPG.1231b572a3fb6029989682@news.pi.se>
Message-ID: <GDVx3.48$W%.12357@typhoon-sf.snfc21.pbi.net>

"E:\local\Python\Tools\idle\idle.pyw" is the default startup path for IDLE.
i.e. The present distribution places IDLE in the Tools\idle directory.
If what you want is the path of the present script that is loaded, it will
be returned by
sys.path. The problem with this, is that you get all the paths :>)
Sam Schulenburg

Svante Kleist <Svante.Kleist at NEMESIS.se> wrote in message
news:MPG.1231b572a3fb6029989682 at news.pi.se...
>
> Dear Pythoneers,
>
> From within a Python script executed in IDLE, how can I get
> its path/file? sys.argv[0] gives me this:
>
> E:\local\Python\Tools\idle\idle.pyw
>
> Thanks for any tip.
>
> / Svante Kleist - NEMESIS systemDesign, Stockholm





From simngolda at voyager.net  Mon Aug  9 22:56:40 1999
From: simngolda at voyager.net (Sim & Golda Zacks)
Date: Mon, 09 Aug 1999 22:56:40 -0400
Subject: Adding href and img to HTMLFMT
References: <7oljd7$1ni@newsops.execpc.com>
Message-ID: <7oo46q$spp@newsops.execpc.com>

It must have been brainfreeze last night. It obviously is a multiple
inheritance using HTMLSingleElement and HTMLElement. Code posted below for
both.

class IMG(HTMLSingleEntry, HTMLElement):
   """ an Image tag"""

Tag_Format="<img%(SRC)s%(ALT)s%(ALIGN)s%(HEIGHT)s%(WIDTH)s%(BORDER)s%(HSPACE)s%(VSPACE)s%(USEMAP)s%(ISMAP)s>"

   end_tag=""

HTML_Attributes=("SRC","ALT","ALIGN","HEIGHT","WIDTH","BORDER","HSPACE","VSPACE","USEMAP","ISMAP")

   def __init__(self,
   SRC=None,
   ALT=None,
   ALIGN=None,
   HEIGHT=None,
   WIDTH=None,
   BORDER=None,
   HSPACE=None,
   VSPACE=None,
   USEMAP=None,
   ISMAP=None
   ):

HTMLElement.__init__(self,SRC=SRC,ALT=ALT,ALIGN=ALIGN,HEIGHT=HEIGHT,WIDTH=WIDTH,BORDER=BORDER,HSPACE=HSPACE,VSPACE=VSPACE,USEMAP=USEMAP,ISMAP=ISMAP)


class A(HTMLSingleEntry, HTMLElement):
   """an HTML hyperlink"""

Tag_Format="<a%(HREF)s%(name)s%(charset)s%(target)s%(rel)s%(rev)s%(accesskey)s%(shape)s%(coords)s%(tabindex)s>"

   end_tag="</a>"

HTML_Attributes=("HREF","charset","name","target","rel","rev","accesskey","shape","coords","tabindex")

   def __init__(self,
      HREF,
      Text,
      name=None,
      charset=None,
      target=None,
           rel=None,
      rev=None,
      accesskey=None,
      shape=None,
      coords=None,
      tabindex=None
      ):
 HTMLElement.__init__(self, Text=Text, HREF=HREF, name=name,
charset=charset, target=target, rel=rel, rev=rev, accesskey=accesskey,
shape=shape, coords=coords,tabindex=tabindex)


Sim & Golda Zacks wrote:

> I am using the HTMLFMT module that came with the INTERNET programming
> with Python and I wanted to add support for links and images but I am
> having a problem. I can't figure out whether to add it as a descendent
> of HTMLElement or of a Seq. The problem is it should be a Sequence
> because There are no properties to it, but the URL is part of the tag
> and I seem to be having a mental block when I try thinking about it. If
> someone could help me with this, I would greatly appreciate it.
>
> Sim




From dwalker at iximd.com  Mon Aug 16 09:19:29 1999
From: dwalker at iximd.com (Damond Walker)
Date: Mon, 16 Aug 1999 13:19:29 GMT
Subject: Question: copy/deepcopy with the curses module?
References: <0eCt3.1775$kW2.100172@typhoon2.gnilink.net> <slrn7res4l.3ma.amk@207-172-146-60.s60.tnt3.ann.va.dialup.rcn.com>
Message-ID: <BdUt3.1876$kW2.128243@typhoon2.gnilink.net>

A.M. Kuchling wrote in message ...
>
>Here's a suggestion (that I haven't tried): create a new window the
>size of the entire screen and render the SIO instance into it; this
>should overwrite all the other information on the screen.  Then
>restore the old display by deleting the new window and refreshing the
>screen.  Something like 'win = stdscr.newwin(0,0) ; ... do stuff with
>win ; del win ; stdscr.refresh()'...
>


    Since I posted the original message I found inch().  I was thinking
about saving the text that is about to be overwritten, and when I'm done
with the current object, just repaint from this buffer.  Throw this code
into my class Dialog (which is nothing more than a screen region if you look
at it from a certain standpoint) and it becomes automagic.
    Unless, of course, the window will take care of all that for me.  I'll
hack it out later tonight.  The buffer method would probably be quite slow.
:)
    There is nothing stopping me from manhandling the SIO instance for while
there are many objects using it, only one uses it at a time.  So it becomes
a matter of cleaning up after myself.

                        Damond





From skip at mojam.com  Sat Aug 21 20:39:26 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sat, 21 Aug 1999 19:39:26 -0500 (CDT)
Subject: Profiler - counting individual lines' exec time
In-Reply-To: <slrn7rtr32.hk.smithy@hoohoo.org>
References: <slrn7rtr32.hk.smithy@hoohoo.org>
Message-ID: <14271.17541.12909.908766@dolphin.mojam.com>

    ?ukasz> I have a loop in my program which loops hundreds of thousands
    ?ukasz> times and I want to know which instructions in it take the
    ?ukasz> most time to execute. Is that possible?

Time per line?  I don't know if there's a module that does that today.  As I 
recall, profile does per-function time measurement.  You might try my trace
module (check http://www.musi-cal.com/~skip/python/ for a pointer).  It
counts the number of times each line is executed.  Helpful, perhaps, but
still not quite what you want.  You could instrument it to collect timing
information as well, but that would make it even slower than it already is.
(Perhaps that's a good Saturday evening activity for me... hmmm... ;-)

Perhaps the best way to measure what you want in a fairly unobtrusive way is
to approximate it.  Set an alarm to ring periodically (every 0.05 or 0.1 s,
let's say).  When the alarm goes off, look at the program's current line
number and allocate the entire time slice to the current line.  Averaged
over a large number of samples, it should approximate the timing behavior of
the program as a whole.

You might be able to modify the python debugger or the profile module to do
this rather easily.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From JamesL at Lugoj.Com  Wed Aug 25 00:24:29 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Tue, 24 Aug 1999 21:24:29 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk>  		 <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu>  		 <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu>  		 <37c126a6$0$813@newsreader.alink.net>  		 <slrn7s2dtk.vsl.kaz@ashi.FootPrints.net> <37c21fae$0$216@nntp1.ba.best.com>
Message-ID: <37C36FFD.7C1501A0@Lugoj.Com>

Xah wrote:
> English has evolved to be extremely
> versatile and useful. It gives you the freedom to express anything in any
> manner or style you want.

Don't know any English; just American <wink>. All that freedom leads to
ambiguity. Great for poetry; shitty for scientific work. So genius, how many
meanings do you think can be formed from the simple spoken phrase "Pretty
little girls school." Note the qualification of "spoken". The ear may hear
"girls," but can't tell if you meant "girl's" or "girls". And you can't
determine it from context. But even if you pick the plural form, I doubt you
could figure out all the meanings. And you are allowed to include only those
meanings that are of high plausibility.

The bottom line is that the English language is just not a good template to
use to design unambiguous languages. (I still don't know about the American
or Canadian languages though....)

> There
> is nothing whatsoever in math symbolism that cannot be expressed better by
> English. For example, a^2+b^2=c^2 can be expressed as "A number multiplied
> by itself plus another number multiplied by itself is equal to a third
> number multiplied by itself". Now, that's the verbose version you are
> thinking of. But look, "Sum of two squared equals one" is a condensed form.

Sorry, could not make sense out of that last sentence. Translated it to
something like 2^2 + ? = 1, which can't be right. So what color is the sky
in your world?

> Further, "Sum two squares is one". Or if you are a power speaker: "s2is1",
> which is more terse than the symbolic form and no information is lost
> because of _context_. (no, pure symbolism cannot do away with context) You
> know what a 'ball' is don't you? But in symbols, it's something like
> a^2+b^2+c^2=d with some elaborate coordinate system and an arcana of
> presumptions.
> 
> I suggest that you go to www.perl.com and read Larry's speeches of perl
> conf. Anyone will do.

I just tried that. That man rambles on and on in an almost incoherent
manner.

> It really depends on who's doing the
> reading. If you are a math guy, then perhaps symbols is your thing. Most of
> us would prefer English. Legibility is purely subjective.

I note you say "Most of us," rather than "I". Who else are you writing for?
I insist that you not include me as an unwitting partner in your absurd
tirade.

> Granted that a new language like math symbolism might have little
> theoretical advantage down the road for some theoretical cleanness,
> precision, or maybe expressiveness, but English has been used for literally
> eons.

Surely you could be more precise than "eons". I'm not sure how long a
literal eon lasts. My dictionary says an eon is "an extremely long,
indefinite period of time; thousands and thousands of years." (Useful hint:
many dictionaries actually include a brief history of the English language.)
But I think I now know where you are coming from; I still haven't figured
out the color of the sky on your world though.

> The inertia is strong, the legacy is golden, and no one can stop it.
> As I said before, it has also evolved into perfection. People will use
> English and people will prefer English over any arcane symbolism. It is
> egregious a crime that math text books are littered with arcane symbols
> instead of plain English. It is why no one in the world likes math.

Euler just hated it. So did Gauss. They are both dead, so obviously it
didn't help them any. And thank God the bible was written in English, like
the good Lord intended!

> 
> One thing with English is that it has not evolved into a computing language
> yet, thus Perl comes to the rescue. Here's a little history lesson for you:
> 
> Larry wanted to process
> some text as his job requires at Jet Propulsion Lab. However, with all
> unix's power and piping shit with sed and awk and baggage but he couldn't do
> it well. Therefore, following the unix tradition, he quickly hacked up his
> own language and put it on the net thinking that he's done human kind a big
> favor. Since unix community consists of largely of morons, they quickly
> picked up Larry's tool without much decision process. Thus, it gradually
> rolls into a snow ball. Perl's success is a direct result of democracy.

Oh; now I get it <wink>. Well, the joke is on me. Don't mind me, I'm a bit
slow on the uptake. I should have read all the way through. And I looked at
your web site. I must admit you write great parody. Anyway, I'll post this
with my comments as I originally started to enter them. Great work.

[Remainder of brilliant work elided.]



From cgw at fnal.gov  Thu Aug 12 18:33:15 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Thu, 12 Aug 1999 22:33:15 GMT
Subject: Help!  $PYTHONPATH trouble
In-Reply-To: <37B33980.A5318181@n-space.org>
References: <37B33980.A5318181@n-space.org>
Message-ID: <14259.19371.42938.565137@buffalo.fnal.gov>

Erik Stewart writes:

 > $ PYTHONPATH="home/possum/test/kwp"
 > $ echo $PYTHONPATH
 > home/possum/test/kwp
 > $ python
 > Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC egcs-2.91.66 19990314/Li
 > on linux-i386
 > Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
 > >>> import sys
 > >>> sys.path
 > ['', '/usr/lib/python1.5/', '/usr/lib/python1.5/test',
 > '/usr/lib/python1.5/plat-linux-i386', '/usr/lib/python1.5/lib-tk',
 > '/usr/lib/python1.5/lib-dynload', '/usr/lib/python1.5/site-packages']

Try exporting the variable.  The way you are setting it above, it becomes a
shell variable, not an environment variable - it's not inherited by
sub-processes.   I think if you do 

>>> import os
>>> printos.environ['PYTHONPATH']

you'll see that PYTHONPATH is in fact not in Python's environment.

Do "export PYTHONPATH" before starting Python and you'll get the
expected results.  Unless of course you have some other completely
different problem ;-)




From thantos at brimstone.mecha  Thu Aug 26 16:37:12 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Thu, 26 Aug 1999 20:37:12 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com> <xlx1zcqv4vj.fsf@gold.cis.ohio-state.edu>  <37C592EC.BFC9FC9@research.att.com>
Message-ID: <slrn7sb965.h05.thantos@brimstone.mecha>

On Thu, 26 Aug 1999 19:18:04 GMT, Garrett G. Hodgson
<garry at research.att.com> wrote:

>thread is that the profitable ones tend to be stated as, "i find
>feature X from language Y really useful.  does/could python
>have this?"

Or, in many cases, 'How can I emulate this feature in Python?'

In the case of, say, Scheme, we've discussed how to impliment
closures.  (I use the creation of a class with a __call__ method to
store local state, myself, on that one.)

>From Haskell/ML, we've discussed Generators and posible syntaxi for
them.  (I, myself, think that list Generators should be added to the
language with syntax like Haskell's, but there are certainly other
ways to do that with the tools that're built in to the object system.)

>From Haskell, we've also discussed things like lazy evaluation and
infinite streams.  (A method of handling infinite streams with lmbda
was just posted recently; I, again, would like to see laziness and
infinite streams built in at C speed; there's just so much fun to
saying [1..] and getting back an infinite list of integers to map on.) 

The Python community is remarkably /open/ to cross-language
fertilization, the very design of the language shows that.  The
problem with this most recent fiasco was not, despite some suggestion,
on the side of the Pythonistas in large amount.  We are, in the main,
flexible enough individually that the idea of someone essentially
suggesting "My way is the Right Way(tm)" is inherently insipid and
obstructive to doing what Pythonistas do best: solving problems.

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From jstevens at basho.fc.hp.com  Fri Aug 13 23:02:38 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:02:38 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132300.RAA06604@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:00 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ce3e@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :> Likewise, because an array and a hash
> :> are, despite their interconvertibility, fundamentally different things,
> :> they *look* different.  Contrast Python's:
> :> 
> :>     b = ["alpha", "beta", "gamma"]
> :>     c = { "fred" : "wilma", "barney" : "betty" }
> :>     print "b nought is", b[0]
> :>     print "c fred is", c["fred"]
> :> 
> :> with Perl's
> :> 
> :>     @b = ("alpha", "beta", "gamma");
> :>     %c = ( "fred", "wilma", "barney", "betty");
> :>     print "b nought is", $b[0];
> :>     print "c fred is", $c{"fred"};
> :
> :Correct.  Python's behavior is preferable.  Think: operator overloading
> :in C++.
> 
> You seem to equate quantifiable benefit with mere opinion.  Please work
> on that.

And you have a tendency to jump to conclusions.  The benefits are
quantifiable.  They are not opinion.  Been there, done the statistics.

> I don't want different things looking the same.

That was the point: they are not different.  Polymorphic.

> It sucks.
> That's an opinion.  I prefer it that way.  That doesn't make it better.

Your opinion is noted.  The benefits of OO, however, are not opinion.

They are indeed quantifiable.

> Objects are *NOT* ipso facto the Answer.

I never said that they were.  They provide a step in the right
direction.

> Bad programmers persist.

Indeed.

John S.




From mal at lemburg.com  Mon Aug  9 05:11:30 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 09 Aug 1999 11:11:30 +0200
Subject: Licenses...
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com> <37AA804D.9A1317C0@houbits.com> <934115482snz@vision25.demon.co.uk> <37ae686e.59511102@news.ricochet.net> <37AE9AC3.34007C4E@lemburg.com>
Message-ID: <37AE9B42.4DF40758@lemburg.com>

phil wrote:
> 
> On Sun, 08 Aug 99 12:31:22 GMT, philh at vision25.demon.co.uk (Phil Hunt)
> wrote:
> 
> >In article <37AA804D.9A1317C0 at houbits.com> nhead at houbits.com "Nigel Head" writes:
> >> There's no licence statement in the distribution at all for now --
> >> my intention is that anyone can use the contents anyway they want,
> >> for non-commercial purposes, providing they keep a reference to the
> >> original source. Oh yes, and I can't take any responsibility for errors.
> >>
> >> Perhaps one of you experienced people can offer some quick advice as to what I
> >> need to do -- do I need to just add a licence.txt file or must I add a header
> >> to every file? What standard text can I use? Is my intent compatible with the
> >> GPL?
> This post raised some concerns for me since i have a Python extension
> that i was going to add to our latest product release. The product is
> not written in Python, nor does it require Python to be running and
> the extension is there because i thought it was cool. However, i have
> to ask whether it is legal to retain this functionality.

Python itself ships with a BSD style license, which means that
you can use it commercially or non-commercially without having
to pay a fee, but instead including a quote of the license in
your products documentation. You are not obliged to include the
source of your extension or Python program when distributing
them just because they happen to use Python.

> >No. The GPL allows anyone to use the software, for commercial or
> >non-commercial purposes. However, if they distribute the software,
> >they must include all source code, including all source code they
> >write which is linked to the software.
> >
> I don't understand this. What is "the software"? As far as i can tell,
> if the software is distributed in binary format (say as a DLL on
> Windoze boxes) then it can be distributed freely with no conditions.
> I've written an extension to a vehicle tracking product that we
> produce that offers a scripting interface. I'm not sure if anyone will
> use this, but i certainly can't distribute it if i have to distribute
> the Python source. Added to which, i don't want to distribute my
> source (except to selected folk) as i don't need the support grief.

If you link against a GPLed product (even if it's done dynamically!),
the whole construction (your product + the GPLed one) automatically
fall under the GPL. This means that you can still sell and distribute
your product, but only if you include the full source code of
the GPLed product
*plus* your product.

The LGPL is a little different: it allows you to link against the
LGPLed product without the need to ship your source code. You still
have to include the source code of the LGPLed product you're linking
against though (including any changes that you have made to it).

If you are using Python extensions written by different authors,
be sure to check their license statement. In case you have any
doubts get in touch with them before using their code. Most 
extensions are under very similar licenses as Python itself,
but some a GPLed and other LGPLed. Also, some are not valid for
commercial use meaning that you'll have to negotiate a special
license with the author to be able to use in your code. This
usually involves paying a license fee.

Hope this clarifies things a little. More infos on GPL and LGPL
should be available on www.gnu.org. If you are still unsure what
is legal and was is not, go on the safe side and contact your
lawyer to get things straight.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   144 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From sisk at mojotoad.com  Sun Aug 15 00:28:27 1999
From: sisk at mojotoad.com (Matt Sisk)
Date: Sun, 15 Aug 1999 04:28:27 +0000
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cf33@cs.colorado.edu>
Message-ID: <37B641EB.1DBC8151@mojotoad.com>

John W. Stevens said:
> Proof, please, that Perl is 100x more popular than Python?
> (To prove this, you must find a random sample of people who
> have used both, then prove that of those who use both equally,
> that 100x as many prefer Perl).

Hmmm.  In order to prove the enthusiasm of those that use 'A',
we must first limit ourselves to those that also use 'B'.

Okay.
Matt



From sasha at ofoto.com  Wed Aug 18 20:23:54 1999
From: sasha at ofoto.com (Alexander Voynow)
Date: Wed, 18 Aug 1999 17:23:54 -0700
Subject: PIL (jpeg optimization)
Message-ID: <U7Iu3.720$Eu4.15766@dfw-read.news.verio.net>

Hi:

    I'm currently using PIL to resize a lot of fairly large jpegs (300 - 500
kbytes).
Im running into odd behavior with regard to the optimize argument for the
save method

myimage.save("mynewfilename.jpg", quality=80, optimize=1)

This seems to work sometimes and not others. Without the optimize arg it
works
without fail. There *seems* to be some correlation with file size. When it
gets too
far above 300k is when it seems to gack, as follows:

    self.draft.save(self.new_path, quality=80, optimize=1)
  File "/usr/local/lib/python1.5/site-packages/PIL/Image.py", line 666, in
save
    SAVE[string.upper(format)](self, fp, filename)
  File "/usr/local/lib/python1.5/site-packages/PIL/JpegImagePlugin.py", line
307, in _save
    ImageFile._save(im, fp, [("jpeg", (0,0)+im.size, 0, rawmode)])
  File "/usr/local/lib/python1.5/site-packages/PIL/ImageFile.py", line 370,
in _save
    raise IOError, "encoder error %d when writing image file" % s
IOError: encoder error -2 when writing image file

Has anybody encountered this? Any thoughts, ideas?


I am using the latest PIL 1.0 final
Python 1.5.2
Solaris

Thanks

Sasha

sasha at ofoto.com





From robin at jessikat.demon.co.uk  Sat Aug 21 04:37:28 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 21 Aug 1999 09:37:28 +0100
Subject: What is a troll by the way? [offtopic]
References: <zelv3.13444$x04.816752@typ11.nn.bcandid.com>
 <37bdeb1b.31915768@news.erols.com> <37BDF3A7.75FEA04F@vision.arc.nasa.gov>
Message-ID: <HSAATIAIVmv3Ewd1@jessikat.demon.co.uk>

In article <37BDF3A7.75FEA04F at vision.arc.nasa.gov>, Chad Netzer
<chad at vision.arc.nasa.gov> writes
>Robert Kern wrote:
>
>> _Troll_ is a USENET term for someone that deliberately posts an
>> inflammatory message simply to stir up argument.
>
>I've found it funny to watch this term evolve over the past few years.
>It started (I believe) as a reference to the fishing term, and used to
>mean "trolling for flames" (ie. hoping to bring flames to the
>discussion), perhaps employing "flame bait" for this purpose.
>
...
presumably the fishing reference is a mis-pronounced form of trawl
although given the number of English dialects troll might be correct
somewhere in that sense.
-- 
Robin Becker



From sams at quinta.com  Mon Aug 30 18:25:26 1999
From: sams at quinta.com (Sam Schulenburg)
Date: Mon, 30 Aug 1999 22:25:26 GMT
Subject: Bit fields
References: <7qeim8$3su$1@nnrp1.deja.com>
Message-ID: <7qf0c8$efi$1@nnrp1.deja.com>

I have written a Python ASPI based interface for SCSI that is usable
under Windows 95/98/NT. Feel free to steal any thing you can use from
the package. http://starship.python.net/~schulenburg


In article <7qeim8$3su$1 at nnrp1.deja.com>,
  archer_sully at my-deja.com wrote:
> Hi,
>
> I'm writing some extensions to set up a general SCSI oriented
> test platform in Python.  Obviously there's a certain amount
> of C work that has to be done, but I would like to do one of the
> following:
>
> A. access bit fields from a C struct that is returned to Python
> as a string
>
> or
>
> B. encapsulate the structs as classes with all of the methods
> implemented in C (this would appear to be possible but the
> doc is mute).
>
> A is preferable, because I wouldn't have to do write a new
> class every time the firmware guys come up with a new struct
> that is made up entirely of bit fields.  B is ok, because at
> least I have emacs to generate the code for me.
>
> Of course, I may be all wet with this question, in which case
> I humbly ask for enlightenment ;-)
>
> TIA,
>
> Archer Sully
> Software Engineer
> OnStream, Inc.
> <http://www.onstream.com>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fdrake at acm.org  Thu Aug 26 16:04:59 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 26 Aug 1999 20:04:59 GMT
Subject: count method only for lists?
In-Reply-To: <37C57DAB.63C7@cs.bham.ac.uk>
References: <37C57DAB.63C7@cs.bham.ac.uk>
Message-ID: <14277.40427.102784.165759@weyr.cnri.reston.va.us>

Stuart Reynolds writes:
 > Does anyone know why there the count method is only defined for mutable
 > sequences (i.e. lists) and not strings and tuples? And for that matter,
 > index too? It seems as though the Python Library Reference (2.1.5.2
 > Mutable Sequence Types) only describes lists and not mutable sequence

  Hmm, what mutable sequences are left out?  Just off-hand, I can't
think of any other than lists in the core language.
  Tuples and strings are definately *not* mutable, though I expect the 
interesting methods will appear in Python 1.6.  Barry Warsaw has
already done some work to implement string methods.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From sjoshi at ingr.com  Fri Aug  6 15:34:12 1999
From: sjoshi at ingr.com (sjoshi at ingr.com)
Date: Fri, 06 Aug 1999 19:34:12 GMT
Subject: Help! Getting used memory
Message-ID: <7ofdbj$m6u$1@nnrp1.deja.com>

Hello
      I have a 'renegade' process which I think doesn't release memory
after it quits. Is there a way to check this user win32 functions. If
so, could someone please let me know how to do this in python. I'm
guessing I might have to use calldll...

thanks
Sunit


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From da at ski.org  Wed Aug  4 14:44:55 1999
From: da at ski.org (David Ascher)
Date: Wed, 4 Aug 1999 11:44:55 -0700 (Pacific Daylight Time)
Subject: CVS snapshots of Python, Zope, etc.
Message-ID: <Pine.WNT.4.04.9908041129370.253-100000@rigoletto.ski.org>

So that those behind firewalls and otherwise without CVS access can access
the latest and buggiest (uh, I mean greatest) versions of Python projects,
I've setup a nightly job to make snaphots of CVS trees.

So far, I have nightly snapshots for:

  Python:   http://starship.python.net/crew/da/pythondists
  Zope:     http://starship.python.net/crew/da/zopedists
  PyOpenGL: http://starship.python.net/crew/da/pyopengldists

(the Zope snapshot is a little out of date because of unrelated network
routing issues).

I'm willing to add new projects to this list, within reasonable disk space
limits.  If you're interested in adding a project, email me at
da at python.net with the relevant CVS information.

--david ascher






From mal at lemburg.com  Fri Aug  6 09:15:14 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 06 Aug 1999 15:15:14 +0200
Subject: [Image-SIG] Proper application of the buffer interface
References: <00b601bee00a$e5d021f0$1101a8c0@bobcat>
Message-ID: <37AADFE2.4CD386BD@lemburg.com>

Mark Hammond wrote:
> 
> M.-A. Lemburg <mal at lemburg.com> wrote in message
> <37AAA7FE.FFE0FD2 at lemburg.com>...
> > Fred L. Drake, Jr. wrote:
> > >
> > > David Ascher writes:
> > >  > Would it also make sense to make struct.pack()
> > >  > return a buffer object
> > >  > instead of a string?
> > >
> > >   Sounds like it to me.  If Guido agrees, I can make the changes.
> >
> > Wait... what's so bad about buffer(struct.pack()) ? Strings already
> > know the buffer interface, so this works just fine already.
> 
> Well, struct.pack() does not return a string - it returns a chunk of
> memory.

The docs say:

pack (fmt, v1, v2, ...) 
     Return a string containing the values v1, v2, ... packed according to the given format. The
     arguments must match the values required by the format exactly. 

> So IMO it makes more sense to say "str(struct.pack())" if you really want a
> string.

While its arguable whether returning a string is the right
thing to do, simply returning a buffer object instead of
a string will certainly break code expecting a string -- you can't
rely on all C APIs using "s#" to parse the return value of
struct.pack(), even though most of them will probably.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   147 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From S.Tian at shu.ac.uk  Fri Aug 13 14:55:11 1999
From: S.Tian at shu.ac.uk (Shicheng)
Date: Fri, 13 Aug 1999 18:55:11 GMT
Subject: Help: sendmail under MS NT
Message-ID: <7p1pma$ak6$1@nnrp1.deja.com>

Under unix platforms, using, for example,

sendmail_cmd = "/usr/lib/sendmail -oi "

to access the SMTP sendmail utility.

Our enquery is: under Windows NT/95/98 platforms,
how to access the "sendmail" utility from within
python code?

Thanks,

Shicheng






Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From dwalker at iximd.com  Fri Aug 13 16:26:40 1999
From: dwalker at iximd.com (Damond Walker)
Date: Fri, 13 Aug 1999 20:26:40 GMT
Subject: Putting code somewhere
References: <7p1tsi$dqe$1@nnrp1.deja.com>
Message-ID: <4c%s3.1113$hv3.57169@typhoon1.gnilink.net>

    I'm wrapping up a simple curses-based interface module (implements
textboxes, buttons, and labels -- told ya it was simple).  It is to be used
for console applications and supports a notion of dialog boxes as well.
Assuming that anyone has any interest in this, where would I go about
posting the code?  python.org?

                 Damond





From tegbert at xmission.xmission.com  Sun Aug  1 13:34:41 1999
From: tegbert at xmission.xmission.com (Tim Egbert)
Date: Sun, 1 Aug 1999 11:34:41 -0600
Subject: RMI and Python
Message-ID: <Pine.GSO.4.10.9908011131490.25758-100000@xmission.xmission.com>

Java is able to do remote method invocation (RMI) which is a neat concept.
It is the basis for Jini, another neat idea.  Does Python have anything
like RMI?

Tim Egbert




From irclark at latveria.castledoom.org  Sun Aug 22 22:37:05 1999
From: irclark at latveria.castledoom.org (Isaac)
Date: 23 Aug 1999 02:37:05 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk>
Message-ID: <slrn7s1cug.1vk.irclark@latveria.castledoom.org>

On Sat, 21 Aug 99 23:56:16 GMT, Phil Hunt <philh at vision25.demon.co.uk> wrote:
>
>
>It looks like version 0.0.1 of Parrot, my text-based GUI builder,
>will be ready within a day or so. It doesn't actually do anything
>useful yet; v0.0.1 is a technology demonstration version, which
>is intended to verify that the basic design is more-or-less
>sensible. Would anyone like to have a look at it, and possibly 
>suggest some improvements? If so, tell me, and I'll put it on
>my website.
>
>Some information about Parrot follows:
>
Very interesting indeed!  I was working on a similar program whose
functionality would fit much of the description you give here.  My
program reads in a description of a gui that's similar in concept
to your par files, and outputs code that implements the gui, in
either Motif, GTK, or the EZWGL toolkit based on the backend chosen.   
Currently the code is limited in the number of supported widgets. 
I haven't tried to output anything but C code, but I don't see
why other procedurally based languages wouldn't work.

I haven't had time to work on my program lately, and the last time I
looked at it, I really didn't like the syntax I'd chosen for the
gui representation.   I'm using PCCTS to implement the parser for the
language.  I used a separate parser for implementing a language 
for describing backends.  It's actually the language implementing part 
that I find the most fun.  I barely learned enough about the various 
toolkits to get the code output part working.

Looks like your seriously working on your program while I just can't
find much time to continue.  (I just started an evening law school
program.)  If you put the code up, I'd be happy to give you all the
feedback I can. 

Isaac



From sp00fD at yahoo.com  Tue Aug 24 13:14:55 1999
From: sp00fD at yahoo.com (sp00fD)
Date: Tue, 24 Aug 1999 17:14:55 GMT
Subject: Converting from perl
Message-ID: <7pujub$75v$1@nnrp1.deja.com>

I'm converting to python from perl and just created my first python
program, it's actually a rewrite of a perl program that I had.  It's
kind of stupid, but I find rewriting things in other languages is one of
the best ways to learn.  What I'd like is for someone to tell me what's
screwed up and what I need to work on.  The majority of the script is
held in a class, but being a newbie to the world of OOP also, I'm not
sure if the way I've handled things is optimal, so _any_ suggestions
would be appreciated, as well as any "Why the hell did you create a
class that way"'s or "You're all screwed up, you need to change
this.."'s.  Anyway, here it is...

<PRE>

#!/usr/bin/env python

import sys
import os

class CopyClass:

        def __init__(self):
                self.hosts = ()
                self.src = None
                self.dst = None

        def usage(self):
                print """Usage: %s -p [preset]
                where preset =
                        AIX     : athena dev detroit la laredo
                                  miami newark sat savannah
                                  seattle titan
                        SUN     : eclipse hercules scdev
                                  scprod sol
                        BACKUP  : athena eclipse hercules
                                  scdev scprod sol xena
                        OTHER   : You choose""" % sys.argv[0]

        def args(self):
                import getopt

                _ok = 1

                try:
                        opts, args = getopt.getopt(sys.argv[1:], 'p:')
                except getopt.error:
                        _ok = 0

                if not _ok or len(args) != 2:
                        self.usage()
                        sys.exit(1)

                for o, a in opts:
                        if o != '-p':
                                self.usage()
                                sys.exit(1)
                        _preset = a

                return _preset

        def files(self):
                if os.path.isfile(sys.argv[3]):
                        self.src = sys.argv[3]

                self.dst = sys.argv[4]

                if self.src and self.dst:
                        return(self.src, self.dst)
                else:
                        self.usage()
                        sys.exit(1)


        def ask_for_hosts(self):
                import string

                self.hosts = string.splitfields(raw_input("Enter hosts:
"))
                return self.hosts

        def get_hosts(self):
                if self.args() == "AIX":
                        self.hosts = ("athena", "dev", "detroit", "la",
\
                                      "laredo", "miami", "newark",
"sat", \
                                      "savannah", "seattle", "titan")
                elif self.args() == "SUN":
                        self.hosts = ("eclipse", "hercules", "scdev", \
                                      "scprod", "sol")
                elif self.args() == "BACKUP":
                        self.hosts = ("eclipse", "hercules", "scdev", \
                                      "scprod", "sol", "athena")
                else:
                        self.hosts = self.ask_for_hosts()

                return self.hosts

        def copy(self):
                _stuff = ()
                _stuff = self.files()

                try:
                        for host in self.get_hosts():
                                print "Sending %s to %s %s" % \
                                (_stuff[0], host, _stuff[1])

                                os.system("rcp %s %s:%s" % \
                                (_stuff[0], host, _stuff[1]))
                except os.error:
                        print "Error during system() function - oops"

c = CopyClass()

c.copy()

</PRE>

Hope that PRE works

thanks


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jbauer at rubic.com  Wed Aug 18 21:47:02 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 19 Aug 1999 01:47:02 GMT
Subject: PIGgies?
References: <19990818171523.A22207@digex.net>
Message-ID: <37BB6216.55DC0E79@rubic.com>

> Am I the only Python enthusiast that has a problem 
> with the term "PIGgies"?

The age of high brow Seriousness is over.  We find
cow sculptures invading Chicago -- even on the ledge 
of the opera house.

Computer languages used to have meaningful, formidable 
names in UPPERCASE:  FORTRAN, ALGOL, PL/I.  Now we
have Python -- named after practitioners of the most
irreverent form of satire since Lenny Bruce.

The age of low brow Whimsicality is upon us.  I see
signs and portends in the distance.

PIGgies take their sense of humor seriously.  Don't mess 
with them unless you want a pie in your face.  ;-)

-Jeff Bauer

P.S. Vote "Silly Party"!



From aaron_watters at my-deja.com  Fri Aug 13 09:07:31 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Fri, 13 Aug 1999 13:07:31 GMT
Subject: designing question
References: <3792D572.7F223EC3@worldnet.fr> <1279735663-4809742@hypernet.com>
Message-ID: <7p15ae$s2q$1@nnrp1.deja.com>

In article <1279735663-4809742 at hypernet.com>,
  gmcm at hypernet.com wrote:
> Pierrette writes:
>
> > I have two questions:
> > 1. The database will hold abouth 50.000 lines in its biggest table,
> > can gadfly handle this without problems, what abouth acces time ??
>
> Gadfly is in-memory, and (to support transactions) has 2 copies of
> the db. If the machines have the memory, access time will be fine.

This is sortof true and sortof deceptive.

The shadow copies of a table share most of their structure
with the committed copies most of the time (ie, they are python
lists that contain pointers to shared structures most of the time)
and shadow copies are only created when a table is updated but
not yet committed.  So really almost always gadfly has only one
copy of almost every data item in the database (unless you update
huge numbers of rows before a commit).

And the answer is...  If you have enough memory gadfly should work
fine.  If you don't you *might* see a serious performance reduction
(unless most of the accesses are via indices in which case it's
hard to say).  Good luck.  Sorry I missed this before.  Look to
the www.egroups.com gadfly-rdbms mailing list for some other
discussions and benchmarks.

  -- Aaron Watters http://www.chordate.com

===

Winston, if you were my husband, I'd poison your tea.
Madame, if you were my wife I'd drink it.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bowman at montana.com  Tue Aug 31 09:36:38 1999
From: bowman at montana.com (bowman)
Date: Tue, 31 Aug 1999 07:36:38 -0600
Subject: Bug in Tkinter?
References: <c75zoz849tq.fsf@pc142.cosc.canterbury.ac.nz> <010401bef398$85ae3e00$f29b12c2@secret.pythonware.com>
Message-ID: <37CBDA66.D406D072@montana.com>

Fredrik Lundh wrote:
> 
> <tre17 at student.canterbury.ac.nz> wrote:
> > permanent reference to StringVar instance, and now it works?
> 
> feature, not bug.  it's related to:
> http://www.python.org/doc/FAQ.html#4.69

I haven't dug into the underlying code, but in those instances where a
variable is not required to make the widgets work like a textvariable in
an Entry, is there any advantage to using a StringVar and calling get()
on it as opposed to just calling directly against the widget?



From kas at maps.magnetic-ink.dk  Sun Aug  8 09:29:04 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Sun, 8 Aug 1999 15:29:04 +0200
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk> <slrn7qq4as.cc2.kas@titan.magnetic-ink.dk> <934111569snz@vision25.demon.co.uk>
Message-ID: <slrn7qr1h0.e10.kas@titan.magnetic-ink.dk>

Phil Hunt wrote:

>>>    #bak.py
>>>    from hhh import *
>>>
>>> [...]
>>> 
>>>    #hhh.py
>>>    from bak import *
>> 
>> Do we have a circular reference here?
> 
> Yes. Does Python not allow this? What's the workaround?

It seems python allows it, however, in this case it is not beneficial.
See, bak tells the interpreter to get all of hhh before proceeding but
as soon as python goes there it is told to go back and get the rest of
bak.  At least thats how I perceive it.

I do not know the details of your code, so I cannot be sure, but a
workaround may have to include the creation of a third module and/or
repositioning the import statement(s) -- it's not a law that they have
to appear at the beginning, but sometimes it gives you a better over-
view.

I would try and reposition the the import statements first, then,
if this doesn't work, creating a third module and rearraniging things
so as to avoid circular references.

Let us know what you find out. :-)


  //Klaus

-- 
???[ Magnetic Ink ]????????????????????????????????????????????????? ><> ???
???[ Delapsus resurgam ]????????????????????????????????????????????????????



From eugene.leitl at lrz.uni-muenchen.de  Tue Aug 10 21:33:03 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Tue, 10 Aug 1999 18:33:03 -0700 (PDT)
Subject: Python GUI
In-Reply-To: <934325065snz@vision25.demon.co.uk>
References: <19990805141904.115.qmail@hotmail.com>
	<37AB0600.869DFD87@NOSPAM.strs.co.uk>
	<934325065snz@vision25.demon.co.uk>
Message-ID: <14256.53967.489242.229268@lrz.de>

Phil Hunt writes:
 
 > I wholeheartedly agree; Python is the best language around, for most
 > tasks that aren't too computationally intensive.

Python plus C goes where Python alone cannot:
http://bifrost.lanl.gov/MD/MD.html 
http://ftp.swig.org/papers/Py97/beazley.html
http://ftp.swig.org/papers/sc96/INDEX.HTM



From faqtori at hotmail.com  Fri Aug 27 01:33:04 1999
From: faqtori at hotmail.com (Factory)
Date: Fri, 27 Aug 1999 15:33:04 +1000
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net> <slrn7s2lb3.k6b.abigail@alexandra.delanet.com> <7putsq$3ql$1@lycaeus.calstatela.edu>
Message-ID: <MPG.122fcf84a811c6099896e8@news.melbpc.org.au>

In article <7putsq$3ql$1 at lycaeus.calstatela.edu>, ramune at bigfoot.com 
says...
> In article <slrn7s2lb3.k6b.abigail at alexandra.delanet.com>,
> Abigail <abigail at delanet.com> wrote:
> [snip amusing rant]
> >Yeah, those math texts from the 16th and 17th century, they are sooooo easy
> >to read, because they lack all the symbols. Not to mention the original
> >Euclid and those other Greeks, who had no symbols at all!
> 
> <GRIN>
> Oh, but they're written with only symbols!  All those confusing greek symbols
> that mathematicians use are the basic building blocks for the language!  Why
> didn't they just use English?
> </GRIN>

  Hmm that might be a bit off, the anchient Greeks had no concept of 
variables, thus you would never see f(x)= 2y in any old greek 
mathematics. IIRC variables were discovered(?) by Al-Jabar sometime 
after.
  Thus the Greeks certainly did not use symbols, and it held back 
mathematics. Not that this applies to python vs. perl.



From thomas at xs4all.nl  Sun Aug 22 08:50:15 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Sun, 22 Aug 1999 14:50:15 +0200
Subject: can some one help a newbie?
In-Reply-To: <935273107snz@vision25.demon.co.uk>
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <7pkke6$lut@dfw-ixnews6.ix.netcom.com> <935227166snz@vision25.demon.co.uk> <7pmv4j$gtt@dfw-ixnews14.ix.netcom.com> <935273107snz@vision25.demon.co.uk>
Message-ID: <19990822145015.A18813@xs4all.nl>

On Sat, Aug 21, 1999 at 10:05:07PM +0000, Phil Hunt wrote:

> The idea that a company would employ someone, who can't really code,
> as a sysadmin is IMO frightening. Unless they don't care whether
> their computer system works, of course.

Yeah, it's frightening, but it happens. Anyone remember the dilbert 'series'
about COBOL ? Pointy-haired-boss sees Bob the Dino, says "Say, aren't you a
COBOL programmer?", Bob replies "No, but i've been told i look like one",
"You're hired".

This currently accurately depicts the (dutch at least) job market for
'people who do something or other with computers.' A g-d(--)less amount of
companies want to join the Hype-err, internet, but have no clue how. They
can employ a company to do it properly, or they can attract a real system
administrator to work part- or full time, or they can hire someone who they
saw had a weird flashy homepage and offered to build likewise for money. The
last option is often the cheapest, even if the homepage ends up below par
and the guy has to learn everything from scratch during workhours. The same
goes for things like network setups, mailservers, etc, for small offices and
the like.

But you're right, they dont really care wether their computer systems work.
Often they wouldn't be able to tell the difference ;)

Two-years-of-ISP-helpdesk-and-sales-experience-does-it-show'ly y'rs, Thomas

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From rainer2207 at my-deja.com  Sun Aug 29 05:52:56 1999
From: rainer2207 at my-deja.com (rainer2207 at my-deja.com)
Date: Sun, 29 Aug 1999 09:52:56 GMT
Subject: Newbie elementary question (please help!!!)
Message-ID: <7qavto$mhb$1@nnrp1.deja.com>

At the moment I have:

class Interface:

      def __init__(self, master):

            ....
            ....

root = Tk()
inter = Interface(root)
root.mainloop()

Can someone please explain to me what 'self' and
'master' are? I'm assuming master is the root
window while self is the instance name of the root
window in this class. I've also seen master=None
can someone explain this to me?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jstevens at bamboo.verinet.com  Fri Aug 13 23:06:08 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: 13 Aug 1999 20:06:08 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <37b4a18f@cs.colorado.edu>
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cf10@cs.colorado.edu>

In comp.lang.perl.misc, you wrote:
>In comp.lang.perl.misc, 
>    "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
>:OTOH, Python is perfect for the OO trained programmer who is part of
>:a team, or where the code will have to be maintained over a long
>:period of time by many different people.
>
>As is Perl.  A sound Perl OO module designed by a competent programmer
>is certainly up to fitting that bill.

Of course.  The question isn't: can it be done?  The question is:
on the average, which is better?

>A CGI monstrosity from a 
>script kiddie isn't, but this is irrelevant.

Actually, they are *VERY* relevant!  Those CGI monstrosities are
Perl, they go into the statistics, and somebody somewhere has to
maintain them, and somebody has to pay to get them maintained.

John S.




From englim at pc.jaring.my  Fri Aug 20 09:44:35 1999
From: englim at pc.jaring.my (Thomas Lim)
Date: Fri, 20 Aug 1999 13:44:35 +0000
Subject: Reading from serial port and dumping to console
References: <37BD20F8.9A984D7E@pc.jaring.my> <37bd325c.257488729@news.u-psud.fr>
Message-ID: <37BD5BC3.988382A8@pc.jaring.my>

Laurent POINTAL wrote:

> On Fri, 20 Aug 1999 09:33:44 +0000, Thomas Lim <englim at pc.jaring.my>
> wrote:
>
> >: t = open('/dev/tty0', 'w')
>
> Try with t = sys.stdout
>
> A+
>
> Laurent.
>
> ---
> Laurent POINTAL - CNRS/LURE - Service Informatique Experiences
> Tel/fax: 01 64 46 82 80 / 01 64 46 41 48
> email  : pointal at lure.u-psud.fr  ou  lpointal at planete.net

To be more specific,  what I was trying to do was to stuff what I read
from the serial port to the keyboard buffer.  Anyway that this can be
done in Python?

Regards,
Thomas.




From alberto at totallygames.com  Mon Aug 16 18:12:24 1999
From: alberto at totallygames.com (Alberto Fonseca)
Date: Mon, 16 Aug 1999 15:12:24 -0700
Subject: Redirecting Python Output
Message-ID: <rrh3pe9goqs55@corp.supernews.com>

Hello, I am embedding Python into my app under Windows.  I would like to
redirect Python's output so that I can just print it myself.  Has anyone
done this?  What is the easiest way to do this?

Thanks,

Alberto
alberto at totallygames.com





From jeremy at cnri.reston.va.us  Thu Aug  5 14:54:52 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 5 Aug 1999 14:54:52 -0400 (EDT)
Subject: rexec question
In-Reply-To: <xz4sievhzv.fsf@prakinf.tu-ilmenau.de>
References: <xz4sievhzv.fsf@prakinf.tu-ilmenau.de>
Message-ID: <14249.56828.918361.13395@bitdiddle.cnri.reston.va.us>

There are two completely seperate namespaces that come into play with
rexec.  The first is the namespace of the "controlling code," i.e. the 
part of your program that created the RExec.  The second is the
namespace of the "controlled code," i.e. the code you're running
inside the RExec.  Names defined in one namespace do not appear in the 
other namespace.

If you want to names 'foo' or 'o' to be visible in the RExec, you need 
to define them there.  For example:

Python 1.5.2+ (#11, Jun 22 1999, 18:01:08) [C] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
startup
>>> import rexec
>>> code = """class foo:
...     def start(self):
...         print "foo running"
... 
... o = foo()
... o.start()
... """
>>> 
>>> filter = rexec.RExec()
>>> filter.r_exec(code)
foo running
>>> 


jeremy



From bruce_dodson at bigfoot.com  Sun Aug 15 18:24:41 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Sun, 15 Aug 1999 19:24:41 -0300
Subject: pythoncom: implementing dispinterfaces
Message-ID: <q7Ht3.64294$jl.40615873@newscontent-01.sprint.ca>

I have a dispinterface defined in a typelib.  I have generated a python
class for it using makepy.  Now I want to implement that same dispinterface
via a Python class.  To do this, is it enough to simply inherit from the
generated class, or should I be making local copies of _dispid_to_func_ etc.
in my own class? (or in each instance?)  The reason I ask is that I was
reading the docstring for the default policy, and I noticed that it may
change those COM attributes in-place as the object is used.

Thanks,
Bruce




From gmcm at hypernet.com  Mon Aug 30 12:26:12 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 30 Aug 1999 11:26:12 -0500
Subject: pop & append
In-Reply-To: <AEjBCBAaxpy3EwFp@jessikat.demon.co.uk>
Message-ID: <1276095990-52118025@hypernet.com>

Robin Becker writes:

> Someone from the Zope list tells me that the list pop and append
> operations are thread safe. Is this the case? 

Yes, from the perspective of the thread doing the append / pop (*).

> How can one tell if
> particular bits of python code are 'atomic'?

If it happens in one op-code, (how's that for a non-answer :-)).

(*) The fact that the append / pop is thread safe doesn't mean that 
it won't screw up another thread which is iterating the list. Such 
iterations should always use a copy (list[:], which is also atomic), 
or protect the list with a Condition (and iterate as fast as 
possible).

- Gordon



From philh at vision25.demon.co.uk  Sat Aug  7 21:28:30 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 08 Aug 99 01:28:30 GMT
Subject: Problem with modules refering to each other
Message-ID: <934075710snz@vision25.demon.co.uk>

I have two python modules, both of which need to refer to classes 
defined in the other one. How do I do this? 

I am currently putting at the start of each module a line of the form:

   from theOtherModule import *

However, this is not working. My modules -- greatly simplified -- look
like this:

   #bak.py
   from hhh import *
   ti = AnHhhClass()
   print ti

and:

   #hhh.py
   from bak import *
   class AnHhhClass: pass

When I run ``python bak.py'', I get the error message 
``NameError: AnHhhClass''.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From stephen at webadmins.com  Sun Aug 29 15:05:11 1999
From: stephen at webadmins.com (Stephen)
Date: Sun, 29 Aug 1999 15:05:11 -0400 (EDT)
Subject: Why do we call python a scripting language?
In-Reply-To: <37c94c0c.1471116@news.netset.com>
Message-ID: <Pine.LNX.4.04.9908291503460.4018-100000@gnr.u2me3.com>

Point taken. 

/***************************************************************\
 * Stephen ("xinu") Klassen -- Unix Systems Administrator      *
 * Website: http://stephen.webadmins.com/   UIN: 19224641 (wk) *
\***************************************************************/

On Sun, 29 Aug 1999 guy_oliver at yahoo.com wrote:

> On Sun, 29 Aug 1999 04:00:14 -0400 (EDT), Stephen
> <stephen at webadmins.com> wrote:
> 
> ><snipped for brevity>
> >'not a real language capable of real programs.'??
> >
> > Are you on crack? Or do you have a different definition of the 
> > word 'real'? Perl and python or both 'real' languages. I can
> > write just about anything from sockets to log parsing in 
> > them with the greatest of ease.
> 
>  You missed the point.  I said 'others' keep suggesting to me that
> python is not appropriate for some job or the other since they hear
> its a scripting language and equate scripting with simple programs.
> Its mostly because their experience with scripts is limited to simple
> things.  I recognize that most scripting languages can do a lot of
> work.  I write just about everything in python, as I said in the post.

[snipped for brevity]




From stig at ii.uib.no  Tue Aug 10 08:12:46 1999
From: stig at ii.uib.no (Stig E. Sandø)
Date: 10 Aug 1999 14:12:46 +0200
Subject: Python in web page
References: <37a70f28.96313821@midpoint>
Message-ID: <slrn7r05pu.818.stig@ara.ii.uib.no>

On Tue, 03 Aug 1999 15:48:40 GMT, Timothy Dixon <tdixon.no at spam.fwi.com> wrote:
>Is it possible to host a Python application in a web page?  TCL has a
>plug-in for IE and Netscape; is there something similar for Python?

You might want to try out ICEBrowser which is released later this
week with JPython-support, which will let you embed the browser in
python code, write plugins for the browser in python or just as a 
replacement for Javascript in your web-pages. (Oh and of course you can
write applets in it as mentioned somewhere else in this thread)

URL: http://www.icesoft.no/

-- 
------------------------------------------------------------------
Stig Erik Sandoe    Institute of Informatics, University of Bergen
stig at ii.uib.no                         http://www.ii.uib.no/~stig/



From donn at u.washington.edu  Thu Aug 19 13:38:08 1999
From: donn at u.washington.edu (Donn Cave)
Date: 19 Aug 1999 17:38:08 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7ro9ns.h6v.jblaine@shell2.shore.net> <37bc2ee5@cs.colorado.edu>
Message-ID: <7phfe0$19so$1@nntp6.u.washington.edu>

Quoth Tom Christiansen <tchrist at mox.perl.com>,
...
| Something the python community might should learn here from the perl
| community is our zealous position on usable documentation.  HTML alone
| sucks, as does putting things only on the web (two different issues).
| That I can't even get "man python" to come up pisses me off more than
| you can imagine.  I strongly feel that you really should consider making
| something easy for CLI people to use just like any other tool or library.
| That's why we have "pod" as base docs, which can convert into HTML,
| text, info, latex, frame, troff (manpages), etc.  It makes it easy to
| run grep or find programs on.  This is very important.  And each module
| automatically installs that module's manpage (and optionally html doc,
| but this is less popular amongst perl folks).  That gets it added to
| the apropos and whatis and man commands automatically.  This is about
| 0.002% of what the MakeMaker thingie in Perl does, but it's the part
| that matters for docs.  I've surely heard Python folks drooling over it
| for a long time.  Also, the CPAN autoinstaller than runs it all for you.
| All very spiffy.

Searchable CLI documentation is a good thing, and I take that as the main
point of your post here, but the way Perl and Tcl dump man pages into man3
isn't something Python needs to emulate on my account.  Man pages could
be disposed in the same name-space-sensitive way we design libraries,
modules, etc., and if anyone feels like working the Python documentation
into man pages I hope that will be considered.

It's no surprise to me that Python documentation doesn't appear as man
pages, anyway, it doesn't seem like a good match for the information
or the community.  Among the alternatives I know of, I'd go with HTML.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From tchrist at mox.perl.com  Thu Aug 19 19:43:06 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 16:43:06 -0700
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <7pi0v5$nvo$2@newshost.accu.uu.nl>
Message-ID: <37bc887a@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    m.faassen at vet.uu.nl (Martijn Faassen) writes:
:Hm, there seems to be only a discussion on whether docs should be in man
:pages or not in response to this posting here.

Yes, that has been disappointing.  

:I thought the posting was neat. It wasn't particularly useful to me,
:but it's neat. It must've taken quite a lot of work. Thanks!

I'm very glad you liked it.  Thank you.

--tom
-- 
Weinberg's Second Law: If builders built buildings the way programmers
write programs, the first woodpecker to come along would destroy civilization.



From squid at panix.com  Tue Aug 24 03:18:37 1999
From: squid at panix.com (Yeoh Yiu)
Date: 24 Aug 1999 03:18:37 -0400
Subject: word1.word2 notation
Message-ID: <oxgbtbxejea.fsf@panix7.panix.com>

I've noticed in various languages, there is a construction
of expressions with a dot in the middle.  It is a construction
because the name with a dot in the middle is never explicitly 
declared, as in 

	int foo.bar = 42;   // doesn't happen this way

but rather foo is declared in one place and method (or some other 
type of entity)  bar was declared somewhere else and foo.bar was
always available implicitly as a latent method (or as something).

What is the name of this technique and the name of methods 
(or variables ?) so constructed ?

Sorry if this is horribly unclear.  Perhaps an example or two would help:

Here's a python example; datfile.readline() is the thing I'm talking about.

>  datafile = open("/tmp/words", "r");
>  line = datafile.readline()

Here's a visual basic example:

> From MS Word VB Help:  SaveAs
> =====
> Saves the specfied document with a new name or format. The arguments for
> this method correspond to the options in the Save As dialog box (File menu).
> 
> Syntax
> 
> expression.SaveAs(FileName, FileFormat, LockComments, Password,
> AddToRecentFiles, WritePassword, ReadOnlyRecommended, EmbedTrueTypeFonts,
> SaveNativePictureFormat, SaveFormsData, SaveAsAOCELetter)
> 
> expression   Required. An expression that returns a Document object.
> 
> FileName   Optional Variant. The name for the document. 
> FileFormat   Optional Variant. The format in which the document is saved.


squid.




From jstevens at basho.fc.hp.com  Fri Aug 13 23:03:29 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:03:29 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132305.RAA06780@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:05 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ce71@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :Polymorphism requires both OO training, and discipline to use, but if
> :used correctly, it is very powerful.
> 
> No disagreements.  
> 
> But I can't help but wonder: Is this how you keep out 99% of the
> accidental programmers, the ones who use Perl?

Who says I keep anybody out of anything?  I teach OO.  Not C++, or
Smalltalk. . . OO.

> If you require OO training
> and discipline, then you set the bar at the gate untenably high.

An opinion.  Duly noted, of course.  But still an opinion.

> Perl remains proudly pedestrian in its roots.

What does that mean?

> It doesn't require a Computer
> Science degree to use.

Neither does Python.  But, why in heaven's name would you talk first
about competence, then state it as a benefit that Perl can be used
by the untrained?

Why, in heaven's name, would you prefer Joe Blow (who has worked for
five years as a butcher) to perform your brain surgery, to a Medical
Doctor with a degree in Neuro Surgery?

> This, too, is a feature.  Formal training is 
> optional.

   "Formal traing is optional"

Now, what defines the difference between "formal" and "informal"
training?  And, how do you figure that OO is more difficult, more
formal than what-ever-it-is that you are talking about?

John S.




From tim_one at email.msn.com  Thu Aug 19 15:33:08 1999
From: tim_one at email.msn.com (tim_one at email.msn.com)
Date: Thu, 19 Aug 1999 19:33:08 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <8E27B3FC4duncanrcpcouk@news.rmplc.co.uk> <37bc4959@cs.colorado.edu>
Message-ID: <7phm5j$cg8$1@nnrp1.deja.com>

[duncan at rcp.co.uk (Duncan Booth)]
> :However I agree it can be confusing that when you
> subscript a string you get another object of type string.

[tchrist at mox.perl.com (Tom Christiansen) wrote:]
> *Tell me* about it!
>
>     >>> s = "fred"
>     >>> print s[0]
>     f
>     >>> print s[0][0]
>     f
>     >>> print s[0][0][0]
>     f
>     >>> print s[0][0][0][0]
>     f
>     >>> print s[0][0][0][0][0]
>     f
>     >>> print s[0][0][0][0][0][0]
>     f

Hmm.  I don't see how this differs from the Perl:

$x = "fred";
print substr($x, 0, 1), "\n";
print substr(substr($x, 0, 1), 0, 1), "\n";
print substr(substr(substr($x, 0, 1), 0, 1), 0, 1), "\n";
etc

except-that-python-isn't-so-verbose<wink>-ly y'rs  - tim


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From csc635 at mizzou.edu  Tue Aug 24 09:57:39 1999
From: csc635 at mizzou.edu (Cummings, Charles S. (UMC-Student))
Date: Tue, 24 Aug 1999 08:57:39 -0500
Subject: Tkinter documentation
Message-ID: <72805CBA8CD4D211A8090060945155F3063525D6@umkc-mail01>

Does anybody know where to get some end-all ultimate doc for Tkinter
besides what's listed on python.org?

Steven Cummings
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990824/62e3a1aa/attachment.html>

From msteed at altiris.com  Wed Aug 11 10:25:13 1999
From: msteed at altiris.com (Mike Steed)
Date: Wed, 11 Aug 1999 08:25:13 -0600
Subject: gdchart-py-0.1
Message-ID: <51CBA4F0B84CD211BBA60060080588D27ABC49@exchange.keylabs.com>

Hi all,

   I found Bruce Verderaime's GDChart library to be just what I needed for a
recent project.  I wrote a Python interface which I am making available.

   See what GDChart can do: http://www.fred.net/brv/chart/
   Python module:
http://members.xoom.com/msteed/software/gdchart-py-0.1.tar.gz

   The Python module includes GDChart (which includes Tom Boutell's gd
library).  I have built and tested the module on Win32 and Linux (x86).
Feedback, etc., is welcome.

Mike Steed





From gmcm at hypernet.com  Thu Aug 12 22:30:02 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 12 Aug 1999 21:30:02 -0500
Subject: Anyone else find Tk akward?
In-Reply-To: <37B34DBB.E5A79760@sprint.ca>
Message-ID: <1277613992-23192690@hypernet.com>

Orlando Vazquez writes:

> Is it just me, or is the way Tk organizes things on the window
> someone weird? I'm just starting to play around with it, and it has
> me puzzled. Wouldnt it be a lot easier to just use a coordinte
> system? For example put a button at (0,0) with length 10 and width
> 20 etc.. Are there any reasons why it _has_ to be this way?

Tk windows are designed to be easily resizable. Once you get used to 
it (which doesn't take long) you'll have a hard time with GUIs which
expect fixed sizes.

- Gordon



From jlrdesign2000 at hotmail.com  Fri Aug  6 23:39:59 1999
From: jlrdesign2000 at hotmail.com (Jake Ramirez)
Date: Fri, 06 Aug 1999 22:39:59 CDT
Subject: how do I?
Message-ID: <19990807033959.79942.qmail@hotmail.com>

Hello out there,
I am quite new at using python and Tcl/tk scripting tools. How do I make 
them work together. Can I get a solid web reference on that? Thanks for the 
info

Jakester


_______________________________________________________________
Get Free Email and Do More On The Web. Visit http://www.msn.com




From bwarsaw at cnri.reston.va.us  Mon Aug  9 13:32:21 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 9 Aug 1999 13:32:21 -0400 (EDT)
Subject: Wash. D.C. Area: DCPIGgies
Message-ID: <14255.4261.58698.664568@anthem.cnri.reston.va.us>

Notice to all Washington DC area Python users:

We are pleased to be resurrecting the DC area Python interest group
meetings, now called DCPIGgies.  We will be having our first meeting
at CNRI in Reston, Virginia on Tuesday August 31, 1999 from 7:30pm to
9:30pm.  The aim is to keep this meeting pretty technical.

To that goal, we're going to have two exciting presentations.  For the
first hour, the Digital Creations guys will be giving a talk on Zope
2.0's unique object model (acquisition, run-time classes,
transactional objects, etc.) including how Python programmers might
use this in their own non-Zope projects.

In the second hour, Andrew Kuchling will be talking about something
XML-ish, probably processing the Open Directory Project's
(www.dmoz.org) RDF dumps.

There will be question and answer periods after each talk, and we may
have an "open phones" going later than 9:30 if there's interest and if
Guido or I am still awake. :)

Pizza, salad, and soda will be provided.  We are asking for a
voluntary $5 donation at the door to cover the costs of food.  Please
RSVP by email to bwarsaw at cnri.reston.va.us.  Let me know how many plan 
to attend and whether you will be eating pizza.

Directions to CNRI can be found on the Web:

    http://www.cnri.reston.va.us/directions.html

See you then!
-Barry



From clgonsal at keeshah.penguinpowered.com  Sat Aug 28 03:08:05 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Sat, 28 Aug 1999 07:08:05 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <slrn7sasvb.397.kzaragoza@kzaragoza.ne.mediaone.net> <slrn7scf7d.pa.clgonsal@keeshah.penguinpowered.com> <m3wvuh8yo3.fsf@pcrm.win.tue.nl>
Message-ID: <slrn7sf2m1.1gv.clgonsal@keeshah.penguinpowered.com>

In comp.lang.python, you wrote:
> clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves) writes:
> 
> > It is possible, but more complicated, to make this a compile time
> > error. This would essentially require a function/procedure
> > distinction with restrictions about how you used return in each one.
> > There would still have to be a runtime error for cases where someone
> > tried to get a result from a procedure, but cases like the above
> > example could be caught at compile-time. (The above, being a
> > function, must return a value. Procedures would have to defined in
> > some other way, like using some new keyword "proc" in place of
> > "def")
> 
> Moreover, this would require different syntax when *calling* the
> function/procedure. Let's say, procedure names start with `&' ?

No it wouldn't. It would just be a runtime error if you called a
procedure but tried to get back a value.

So there would be a compile-time error if you incorrectly returned a
value from a procedure, or didn't return a value from a function. It
would be a runtime error if you called a procedure and tried to get a
value from it. So procedure calls would be illegal in expressions, but
this could only be validated at runtime, much like Python currently has
to check to see if that 'function' you're calling is really a function
(or a callable object) after all...

To really tighten the screws, the grammar could be altered so that
expressions are no longer a type of statement. This would prevent people
from calling functions and ignoring the result. 

Then a new keyword could be added which would work much like Modula-3's
'EVAL' (Modula-3 keywords are uppercase, it's not my fault...). Since
there's already a fairly popular function with the name 'eval', a good
name for this new keyword in Python might be 'discard'. Then you could
type:

    discard foo(6)

This would call the function foo, and discard the result. This would
allow you to call functions for their side-effects, and ignore the
result without having to use a dummy variable. It would force people to
do so explicitly, though. The discard keyword would actually take an
expression as its "argument". The idea is that unadorned expressions
statement would no longer be legal, but there would be "procedure call"
and "discard expression" statements to replace them.

I suppose that really is a "different syntax when calling" though. Calls
in a statement context would have to be procedure calls, calls in an
expression context would have to be function calls. Violating either of
these conditions would result in a runtime error.

Incidently, I am aware of the usefulness of being able to type
expressions interactively to get the result. The above modification does
*not* imply the removal of this very useful feature from Python.

I'm only semi-serious about the discard keyword idea though, actually.
Many people would probably hate it since it seems "too strict". I
certainly wouldn't mind it myself though... It would give me more
confidence in the correctness of my code, and the code of others. That
Python tends to be strict in ways which encourage good coding practices
is one of the reasons I like it.

I really *would* like to see a stronger distinction between procedures
and functions though. I really don't think it's right for a Python
function to implicitly return None if I just say "return" or fall off
the end of the definition. 

At the very least I think it should be an error to try and get a return
value from a function that does this. I think it would be better if
there was a separate class of callable objects (procedures) that don't
allow the return of a value, and that function definitions should
require that leaving the function always be done by explicitly returning
some value.

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From puetsch at gmx.de  Wed Aug  4 04:53:20 1999
From: puetsch at gmx.de (Felix Puetsch)
Date: 4 Aug 1999 08:53:20 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <7o78fb$blb$2@chief.death.de> <slrn7qec6g.oet.kas@titan.magnetic-ink.dk>
Message-ID: <7o8v20$rl5$2@chief.death.de>

kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup) wrote in
<slrn7qec6g.oet.kas at titan.magnetic-ink.dk>: 

>Felix Puetsch wrote:
>
>>> <BODY>
>>> {
>>>   print n*3
>>> }
>>> </BODY>
>> 
>> How would you make a web page display this message, e.g.? I think
>> you should at least define environments for Python code. ('<!--'
>> etc.) 
>
>I think that's what Ian tried to imply with the {} pair
Still rests the question how to print {...} on a web page.

>I didn't
>perceive his article as an implementaion, rather as an idea to be
>elaborated on. 
Okay, you're right.

-- 
Felix P?tsch



From spamfranke at bigfoot.de  Wed Aug 18 13:46:55 1999
From: spamfranke at bigfoot.de (Stefan Franke)
Date: Wed, 18 Aug 1999 17:46:55 GMT
Subject: representation of unbound methods - strange behaviour
Message-ID: <37bbecf1.100354231@news.omnilink.de>

Why does this happen:

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> class A: pass
>>> def f(): pass
>>> A.f = f
>>> A.g = f
>>> A.f
<unbound method A.f>
>>> A.g
<unbound method A.f>
                ^^^

This is independent of the assignment order of f and g:

>>> class A: pass
>>> def f(): pass
>>> A.g = f        # swapped
>>> A.f = f        # order
>>> A.f
<unbound method A.f>
>>> A.g
<unbound method A.f>

Method caching?
Stefan




From never at mail.matav.hu  Sat Aug 21 07:42:48 1999
From: never at mail.matav.hu (Hever Zsolt)
Date: Sat, 21 Aug 1999 13:42:48 +0200
Subject: NEWBIE:Need Help with "Path"
Message-ID: <199908211143.HAA04811@python.org>

> Autoexec.bat
> and this is literally the path:
> path %path%;C:\Perl\bin;path=C:\PROGRA~1\Python\;%path%
path %path%;C:\Perl\bin;C:\PROGRA~1\Python
Try this and restart your computer! 

Zsolt

< http://www.cab.u-szeged.hu/~h532851/python.htm >




From banderson at boi.hp.com  Mon Aug  2 15:31:44 1999
From: banderson at boi.hp.com (Bill Anderson)
Date: Mon, 02 Aug 1999 13:31:44 -0600
Subject: Python and Red Hat Linux 6
References: <379CD547.22A10A50@postoffice.pacbell.net> <379CD5C2.5BA096A2@mail.danbbs.dk> <7nirk7$icq$1@oak.prod.itd.earthlink.net>
Message-ID: <37A5F220.B83EEF99@boi.hp.com>

Spud wrote:
> 
> Yeah, using #!/bin/env/python should help, but Red Hat should support
> pound-bang (#!) anyway. You might want to check in with tech support.
> 


Uhmm ... it is not RedHat, but a shell thing, and if #! didn't work,
neither would #!/bin/env/python.

-- 
Bill Anderson 					Linux/Unix Administrator, Security Analyst
ESSD (ARC)						banderson at boi.hp.com
My opinions are just that; _my_ opinions.



From bruno.lienard at vnumail.com  Mon Aug 30 15:10:37 1999
From: bruno.lienard at vnumail.com (Bruno Liénard)
Date: Mon, 30 Aug 1999 21:10:37 +0200
Subject: Spread/Worksheet
References: <37CA8EAB.5DD1314@uniserve.com>
Message-ID: <7qeks8$pd5$1@news4.isdnet.net>

Try WxPython, there is such a widget.

Bruno


Justin Muir <jkmuir at uniserve.com> a ?crit dans le message :
37CA8EAB.5DD1314 at uniserve.com...
> Hello all.  I'm looking for a widget that has a spreadsheet(row,col)
> type scrollable interface.  Anyone seen anything?
>
>
> Justin.
>





From godzilla at netmeg.net  Mon Aug  9 23:09:25 1999
From: godzilla at netmeg.net (Les Schaffer)
Date: Tue, 10 Aug 1999 03:09:25 GMT
Subject: looking for py/database ideas
References: <m2so5slkdj.fsf@netmeg.net> <kNKr3.879$gO1.33509@news2.giganews.com>
Message-ID: <m2n1w0l4al.fsf@netmeg.net>

Chris Browne wrote: 

> I've got some material at the URL below ... [snip]
> Rather than being Python-centric, it's somewhat Linux-centric.

thanks. i'll take a look through what you have. 

not sure my friend is so much interested in linux vs. NT/9X. but he
may be interested in an alternative approach to the database solution,
and if that takes him away from Windows, "so be it".

> By the way, the term "integrable" is often used by mathematicians, but
> almost never by "systems integrators."

whats a "systems integrator"? i've never met one so i dont know how
they talk.

is integratable a word?

les



From conversy at lri.fr  Fri Aug 20 11:07:04 1999
From: conversy at lri.fr (Stephane Conversy)
Date: Fri, 20 Aug 1999 17:07:04 +0200
Subject: Tkinter and popup menu
Message-ID: <37BD6F18.34C499AB@lri.fr>

Hi.

I want a popup menu appear when clicking with the right button on a
canvas
(much like a contextual menu).

how to do this ?

    stef

--
St?phane Conversy
http://www-ihm.lri.fr/~conversy/
mailto:conversy at lri.fr






From R.Hooft at EuroMail.com  Fri Aug 20 03:34:28 1999
From: R.Hooft at EuroMail.com (Rob Hooft)
Date: 20 Aug 1999 09:34:28 +0200
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> 	<14265.29899.470021.309355@dolphin.mojam.com> 	<14265.47748.241386.291144@buffalo.fnal.gov> <14265.49248.386119.227023@bitdiddle.cnri.reston.va.us>
Message-ID: <m3k8qqj46z.fsf@devel.no.rwd.dinet>

>>>>> "JH" == Jeremy Hylton <jeremy at cnri.reston.va.us> writes:

 >> The "hylton" method favors even permutations over odd ones, tends
 >> to produce increasing sequences, and distributes the variance
 >> unequally, apparently peaking near the beginning (1/e of the way
 >> through?).  I haven't quite fully analyzed the reasons for this
 >> yet...

 JH> I expect that the reason its tends to produce increasing
 JH> sequences is that you started with a sorted sequence; if you
 JH> reverse the sequence, it will probably tend towards decreasing
 JH> sequences.  I think this follows from the quicksort approach.
 JH> Quicksort is going to divide each list into two sublists where
 JH> the each element of the first sublist is less than or equal to
 JH> each element of the second sublist.

 JH> Since the comparison function is random, the sort is just going
 JH> to exchange a random number of elements around the pivot point.
 JH> Then quicksort is going to be applied to each sublist, so no
 JH> element from the second sublist is ever going to be moved to the
 JH> other side of the original pivot point.  The end result is that
 JH> the shufflesort method will often leave elements close to where
 JH> they were originally.

You can improve the "randomness" performance significantly (still not
making it good) by making the random process never return 0. It is too
complicated for me to go through the quicksort algorithm, but you can
easily see that returning one of -1,0,1 makes a simple sort algorithm
leave 2/3 of the orders the same, and swap only 1/3. This gives a bias
towards the original order.

Rob
-- 
=====   R.Hooft at EuroMail.net   http://www.xs4all.nl/~hooft/rob/  =====
=====   R&D, Nonius BV, Delft  http://www.nonius.nl/             =====
===== PGPid 0xFA19277D ========================== Use Linux! =========



From fredrik at pythonware.com  Mon Aug 16 03:57:12 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 16 Aug 1999 07:57:12 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <37b5a8e0@cs.colorado.edu> <7p7d22$1hh$1@holly.csv.warwick.ac.uk>
Message-ID: <004f01bee7bc$f9945b70$f29b12c2@secret.pythonware.com>

Andrew Clover <esuzm at primrose.csv.warwick.ac.uk> wrote:
> I'd like a syntax like:
> 
>   x.+(1)
> 
> With the idea that x is an object (of type int) with a '+' method that
> adds stuff to it. It'd be nicer without the brackets though.

mutable integers?  now that's a really cool feature! (known
from Fortran and Tcl, among others. I'm pretty sure Perl
has them too ;-).

</F>




From morse at harborcom.net  Fri Aug  6 08:04:34 1999
From: morse at harborcom.net (Kevin Dahlhausen)
Date: Fri, 06 Aug 1999 12:04:34 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <19990805094020.A1838753@vislab.epa.gov>
Message-ID: <37aacdf9.324492916@news.oh.verio.com>

There is a TK/TCL builder called 'SpecTCL.' that has a Python module called
"SpecPython."  I don't have the URL though.   It works very well, if you like
tkinter.

Under linux, there is a wrapper for fltk that translates the data files from the
FLTK gui builder to Python.  This one is very much a work in progress though.
It should also work under Windows, but that is still theory at this point.
http://fltk.netpedia.net

Another option is to use a C++ lib that has a gui builder, and place selected
callback functions into a Python extension  module.  I tried this once for an
earlier version of FLTK.  It worked, and did allow lots of control over the gui.
but there was a lot of programmer overhead in creating and tracking the callback
functions.



Randall Hopper <aa8vb at yahoo.com> wrote:
> |Is there a Python GUI builder out there anywhere?




From brond at abinitio.com  Fri Aug 20 12:39:28 1999
From: brond at abinitio.com (Brond Larson)
Date: Fri, 20 Aug 1999 12:39:28 -0400
Subject: python threads on solaris -- how to get pthread semantics
Message-ID: <37BD84C0.3CCC7D1@abinitio.com>

Is there a supported way to get the python threads packages to build
with 
posix threads on solaris?  By default, the --with-threads configure
switch seems to use native solaris threads.  The difference between
native 
threads and pthreads makes a big difference for fork(2) semantics and
possibly 
other things, and I'd like to be using pthreads on whatever platforms
they are 
available.  By playing with defines and changing link lines I've rebuilt
python to 
use pthreads on solaris, and the tests at least pass, so it doesn't seem
like an 
obviously bad thing to do.  

Can anybody offer insight as to 
(1) why the default is not posix threads whenever they are available on
a platform?  In 
    particular are there known problems on solaris with using posix for
your threads? 
(2) is there an approved way to get posix thread semantics that I'm
missing 
   (unless hacking makefiles is the approved way :-)



From kuncej at mail.conservation.state.mo.us  Thu Aug 26 10:55:25 1999
From: kuncej at mail.conservation.state.mo.us (Jeffrey Kunce)
Date: Thu, 26 Aug 1999 09:55:25 -0500
Subject: MSDOS window
Message-ID: <s7c50f37.003@mail.conservation.state.mo.us>

Launching python with the "-i" command-line option
will keep the window open when the script is
finished. 

I like to define a new action for .py files called "test"
For "Application used to perform action", I specify:
   "C:\Program Files\Python\python.exe" -i "%1"
If I right-click a python script, and select "test", the
script will run, but the window will stay open.

This also works for debugging: when the script stops
running (normally or with an error), you can examine
variables and execute statements interactively from
the command line. You can also run the python
debugger (pdb) on the results.

  --Jeff


>>> <henry299 at my-deja.com> 08/25/99 10:21PM >>>
Hi, i'm new at this. I just downloaded version
1.5.2 for windows 95. When i double click on a
program in the windows explorer, (example.py
which is  print "This is some example text") an
MS-DOS window opens and the text is displayed,
but the window closes by itself in less than a
second. Am i doing something wrong, and is there
a way to keep the popup window open? Any help
would be greatly appreciated.

Thanks in advance,

Matt


Sent via Deja.com http://www.deja.com/ 
Share what you know. Learn what you don't.




From m.faassen at vet.uu.nl  Thu Aug 19 18:37:25 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 19 Aug 1999 22:37:25 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
Message-ID: <7pi0v5$nvo$2@newshost.accu.uu.nl>

Tom Christiansen <tchrist at mox.perl.com> wrote:
> Per your collective requests, here is one of the documents
> with my collective observations from learning Python.

Hm, there seems to be only a discussion on whether docs should be in man
pages or not in response to this posting here.

I thought the posting was neat. It wasn't particularly useful to me,
but it's neat. It must've taken quite a lot of work. Thanks!

Regards,

Martijn




From Tim.Teulings at materna.de  Thu Aug 26 10:00:36 1999
From: Tim.Teulings at materna.de (Tim Teulings)
Date: 26 Aug 1999 14:00:36 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7pp517$dce$1@nnrp1.deja.com> <slrn7s8fgl.d6.grant@grante.comtrol.com>
Message-ID: <slrn7sai44.6j.Tim.Teulings@is_de633.materna.de>

Hallo!

>The text-based specification langauge methodology is also used
>by the GUI package for Modula-3.  The interface is defined in a
>text file by building up composits of controls, containers
>(vertical and horizontal) and glue (hbox,vbox,glue: very
>TeX-like -- with Lisp s-expressions as the syntax).  The result
>can be parsed at compile-time and stored in the executable, or
>parsed at run-time.  Or, if you're feeling particularly
>masochistic, you can generate the s-expressions on-the-fly in
>your program and parse them as you go.

And there is MUI (for the Amiga) and VisualOberon
(http://www.ping.de/sites/edge/VisualOberon.html) for X11/Windows
that use similar layouting concepts.

--
Gru?...
       Tim.



From irclark at latveria.castledoom.org  Tue Aug 24 23:21:08 1999
From: irclark at latveria.castledoom.org (Isaac)
Date: 25 Aug 1999 03:21:08 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <14274.51855.281578.240389@dolphin.mojam.com> <slrn7s6mfo.68l.cbbrowne@godel.brownes.org>
Message-ID: <slrn7s6o94.im.irclark@latveria.castledoom.org>

On Wed, 25 Aug 1999 03:02:30 GMT, Christopher B. Browne 
<cbbrowne at news.brownes.org> wrote:
>
>This of course has the classic problem that
>
>  You can only usefully include things in "Parrot" that are
>  expressible in all of the GUI systems, thereby reducing Parrot to a
>  Lowest Common Denominator.
>
>That's not 100% true; there *could* be Parrot features that would be
>ignored on platforms that did not support them.  Some danger lies down
>that road...
>

As long as they weren't silently ignored, it wouldn't really be dangerous.
The least common denominator for say GTK and motif is actually fairly
useful. 

What really makes this kind of tool hard to write is the differing 
component placement algorithms from one toolkit to the next.   You
probably won't easily be able to support all of the options for a
every toolkit.  Phil's syntax for parrot suggests a certain type
of geometry manager.  Toolkits that won't support that manager
will probably be the most difficult to support with Parrot.

Isaac



From phd at emerald.netskate.ru  Tue Aug 10 06:47:27 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Tue, 10 Aug 1999 14:47:27 +0400 (MSD)
Subject: Hidden CGI fields in HTML
In-Reply-To: <37B0017F.3569CF74@strs.co.uk>
Message-ID: <Pine.LNX.4.04.9908101441180.11952-100000@emerald.netskate.ru>

On Tue, 10 Aug 1999, Ian Clarke wrote:
> On a related issue, you are clearly experienced in using CGI with
> Python.  I couldn't understand why cgi.FieldStorage returned a
> dictionary which contained MiniFieldStorage objects (which must be
> accessed using the .value field) rather than just the data itself.  This
> would be much more convenient as I see it.  Any ideas?

   There are, actually, some more issues on using FieldStorage, for
example, you cannot ask for non-existing field:

v = form["nofield"].value

   will raise KeyError. And there is no way to emulate dict.get()...

   Oh, yes, I think I know why there is .value! What if your form has more
than one value for some key? Like in the following GET:

http://my.server/my/cgi/prog.py?city=Moscow&city=Paris&city=London

   How do you get the value? How do you recognize thera are many values?
Currently, you try to ask for form["city"], if the field exists you try to
ask it for .value, but if the .value does not exists, you know it is a
list:

   for f in form["city"]:
      print f.value

> Cheers,
> 
> Ian.

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From skip at mojam.com  Mon Aug 23 14:57:35 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 23 Aug 1999 13:57:35 -0500
Subject: Forgiving marshal module
Message-ID: <199908231857.NAA20350@dolphin.mojam.com>

Guido wrote a version of marshal in Python for an early version of JPython.
(I think it may have been supplanted by a Java version by now.)  I modified
its Unmarshaller class to be forgiving in the face of truncated marshal
strings.  I call it marshalp.py so as to allow it to be imported alongside
the real thing.  Perhaps someone else will find it useful in an emergency.

half-a-loaf-is-better-than-none-ly y'rs...

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...


"""Marshal module written in Python.

This module will marshal code objects if the new module is available.
Performance or careful error checking is not an issue.

In the face of truncated input, the Unmarshaller object will (should) return
what it can instead of simply raising an exception.  While this generally
not the desired behavior, it can be helpful when trying to recover the
contents of damaged files.

"""

import sys, string
try:
    import cStringIO
    StringIO = cStringIO
except ImportError:
    import StringIO

from types import *
try:
    import new
except ImportError:
    new = None

TYPE_NULL     = '0'
TYPE_NONE     = 'N'
TYPE_ELLIPSIS = '.'
TYPE_INT      = 'i'
TYPE_INT64    = 'I'
TYPE_FLOAT    = 'f'
TYPE_COMPLEX  = 'x'
TYPE_LONG     = 'l'
TYPE_STRING   = 's'
TYPE_TUPLE    = '('
TYPE_LIST     = '['
TYPE_DICT     = '{'
TYPE_CODE     = 'c'
TYPE_UNKNOWN  = '?'


class Marshaller:

    dispatch = {}

    def __init__(self, f):
	self.f = f

    def dump(self, x):
	self.dispatch[type(x)](self, x)

    def w_long64(self, x):
	self.w_long(x)
	self.w_long(x>>32)

    def w_long(self, x):
	write = self.f.write
	write(chr((x)     & 0xff))
	write(chr((x>> 8) & 0xff))
	write(chr((x>>16) & 0xff))
	write(chr((x>>24) & 0xff))

    def w_short(self, x):
	write = self.f.write
	write(chr((x)     & 0xff))
	write(chr((x>> 8) & 0xff))

    def dump_none(self, x):
	self.f.write(TYPE_NONE)
    dispatch[NoneType] = dump_none

    def dump_ellipsis(self, x):
	self.f.write(TYPE_ELLIPSIS)
    try:
	dispatch[EllipsisType] = dump_ellipsis
    except NameError:
	pass

    def dump_int(self, x):
	y = x>>31
	if y and y != -1:
	    self.f.write(TYPE_INT64)
	    self.w_long64(x)
	else:
	    self.f.write(TYPE_INT)
	    self.w_long(x)
    dispatch[IntType] = dump_int

    def dump_long(self, x):
	self.f.write(TYPE_LONG)
	sign = 1
	if x < 0:
	    sign = -1
	    x = -x
	digits = []
	while x:
	    digits.append(x & 0x7FFF)
	    x = x>>15
	self.w_long(len(digits) * sign)
	for d in digits:
	    self.w_short(d)
    dispatch[LongType] = dump_long

    def dump_float(self, x):
	write = self.f.write
	write(TYPE_FLOAT)
	s = `x`
	write(chr(len(s)))
	write(s)
    dispatch[FloatType] = dump_float

    def dump_complex(self, x):
	write = self.f.write
	write(TYPE_COMPLEX)
	s = `x.real`
	write(chr(len(s)))
	write(s)
	s = `x.imag`
	write(chr(len(s)))
	write(s)
    try:
	dispatch[ComplexType] = dump_complex
    except NameError:
	pass

    def dump_string(self, x):
	self.f.write(TYPE_STRING)
	self.w_long(len(x))
	self.f.write(x)
    dispatch[StringType] = dump_string

    def dump_tuple(self, x):
	self.f.write(TYPE_TUPLE)
	self.w_long(len(x))
	for item in x:
	    self.dump(item)
    dispatch[TupleType] = dump_tuple

    def dump_list(self, x):
	self.f.write(TYPE_LIST)
	self.w_long(len(x))
	for item in x:
	    self.dump(item)
    dispatch[ListType] = dump_list

    def dump_dict(self, x):
	self.f.write(TYPE_DICT)
	for key, value in x.items():
	    self.dump(key)
	    self.dump(value)
	self.f.write(TYPE_NULL)
    dispatch[DictionaryType] = dump_dict

    def dump_code(self, x):
	self.f.write(TYPE_CODE)
	self.w_short(x.co_argcount)
	self.w_short(x.co_nlocals)
	self.w_short(x.co_stacksize)
	self.w_short(x.co_flags)
	self.dump(x.co_code)
	self.dump(x.co_consts)
	self.dump(x.co_names)
	self.dump(x.co_varnames)
	self.dump(x.co_filename)
	self.dump(x.co_name)
	self.w_short(x.co_firstlineno)
	self.dump(x.co_lnotab)
    try:
	dispatch[CodeType] = dump_code
    except NameError:
	pass


class NULL:
    pass

class Unmarshaller:

    dispatch = {}

    def __init__(self, f):
	self.f = f
	    
    def read(self,n):
	bytes = self.f.read(n)
	if len(bytes) < n: raise EOFError
	return bytes

    def load(self):
	c = self.read(1)
	return self.dispatch[c](self)

    def r_short(self):
	read = self.read
	lo = ord(read(1))
	hi = ord(read(1))
	x = lo | (hi<<8)
	if x & 0x8000:
	    x = x - 0x10000
	return x

    def r_long(self):
	read = self.read
	a = ord(read(1))
	b = ord(read(1))
	c = ord(read(1))
	d = ord(read(1))
	x = a | (b<<8) | (c<<16) | (d<<24)
	if x & 0x80000000 and x > 0:
	    x = string.atoi(x - 0x100000000L)
	return x

    def r_long64(self):
	a = self.r_long()
	b = self.r_long()
	return a | (b<<32)

    def load_null(self):
	return NULL
    dispatch[TYPE_NULL] = load_null

    def load_none(self):
	return None
    dispatch[TYPE_NONE] = load_none

    def load_ellipsis(self):
	return EllipsisType
    dispatch[TYPE_ELLIPSIS] = load_ellipsis

    def load_int(self):
	return self.r_long()
    dispatch[TYPE_INT] = load_int

    def load_int64(self):
	return self.r_long64()
    dispatch[TYPE_INT64] = load_int64

    def load_long(self):
	size = self.r_long()
	sign = 1
	if size < 0:
	    sign = -1
	    size = -size
	x = 0L
	for i in range(size):
	    d = self.r_short()
	    x = x | (d<<(i*15L))
	return x * sign
    dispatch[TYPE_LONG] = load_long

    def load_float(self):
	n = ord(self.read(1))
	s = self.read(n)
	return string.atof(s)
    dispatch[TYPE_FLOAT] = load_float

    def load_complex(self):
	n = ord(self.read(1))
	s = self.read(n)
	real = float(s)
	n = ord(self.read(1))
	s = self.read(n)
	imag = float(s)
	return complex(real, imag)
    dispatch[TYPE_COMPLEX] = load_complex

    def load_string(self):
	n = self.r_long()
	return self.read(n)
    dispatch[TYPE_STRING] = load_string

    def load_tuple(self):
	return tuple(self.load_list())
    dispatch[TYPE_TUPLE] = load_tuple

    def load_list(self):
	n = self.r_long()
	list = []
	for i in range(n):
	    try:
		list.append(self.load())
	    except EOFError:
		sys.stderr.write("EOF on input ... list incomplete\n")
		break
	return list
    dispatch[TYPE_LIST] = load_list

    def load_dict(self):
	d = {}
	while 1:
	    try:
		key = self.load()
	    except EOFError:
		sys.stderr.write("EOF on input ... dict incomplete\n")
		break
	    if key is NULL:
		break
	    try:
		value = self.load()
	    except EOFError:
		sys.stderr.write("EOF on input ... dict incomplete\n")
		break
	    d[key] = value
	return d
    dispatch[TYPE_DICT] = load_dict

    def load_code(self):
	try:
	    argcount = self.r_short()
	    nlocals = self.r_short()
	    stacksize = self.r_short()
	    flags = self.r_short()
	    code = self.load()
	    consts = self.load()
	    names = self.load()
	    varnames = self.load()
	    filename = self.load()
	    name = self.load()
	    firstlineno = self.r_short()
	    lnotab = self.load()
	except EOFError:
	    sys.stderr.write("EOF on input ... broken code object\n")
	    return NULL

	if not new:
	    raise RuntimeError, "can't unmarshal code objects; no 'new' module"
	return new.code(argcount, nlocals, stacksize, flags, code, consts,
			names, varnames, filename, name, firstlineno, lnotab)
    dispatch[TYPE_CODE] = load_code


def dump(x, f):
    Marshaller(f).dump(x)

def load(f):
    return Unmarshaller(f).load()

def dumps(x):
    f = StringIO.StringIO()
    dump(x, f)
    return f.getvalue()

def loads(s):
    f = StringIO.StringIO(s)
    return load(f)



From davidopp at megsinet.net  Thu Aug 26 23:13:39 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Thu, 26 Aug 1999 23:13:39 -0400
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu>
		<7p49l5$uv7$1@nnrp1.deja.com>
		<37b634a2@cs.colorado.edu>
		<FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net>
		<37C552DF.9F90BA36@megsinet.net>
		<slrn7sblvo.p96.wtanksle@dolphin.openprojects.net> <14277.63776.733073.298192@dolphin.mojam.com>
Message-ID: <37C60262.C6771B75@megsinet.net>

Where can I find info on Minotaur?

David O.


Skip Montanaro wrote:

>     David> That being said, are there ways to integrate Python and Perl to
>     David> do things that neither one can do alone?!
>
>     William> Minotaur is an attempt to do exactly that -- they've got
>     William> Python, Perl, Tcl, Ruby, Forth, and several other languages all
>     William> glued together and all callable from each other, with good data
>     William> interchange.
>
> In a probably less formal way, XML-RPC helps us glue a Perl front end
> (apache+mod_perl+Mason) onto Musi-Cal's database which is written in Python.
>
> Skip Montanaro | http://www.mojam.com/
> skip at mojam.com | http://www.musi-cal.com/~skip/
> 847-971-7098   | Python: Programming the way Guido indented...




From ron at library.ucsf.edu  Tue Aug 10 18:19:16 1999
From: ron at library.ucsf.edu (Ron Klatchko)
Date: Tue, 10 Aug 1999 15:19:16 -0700
Subject: Hidden CGI fields in HTML
References: <37A94EDA.81644EF3@NOSPAM.strs.co.uk>
Message-ID: <37B0A564.E117EB8F@library.ucsf.edu>

Ian Clarke wrote:
> I am using CGI with Python and am trying to find a way that I can
> generate the HTML for a form such that one of the INPUT controls will
> have a fixed value (set when the page is generated by a previous CGI
> script), that CANNOT be edited by the user (or prefereably not even
> seen) - but will be passed to the next CGI script.

That depends on your level of paranoia.  As many people have pointed
out, you can use a hidden field but that only hides at the HTML display
level.  Anyone who wanted to hack that could save the HTML, modify the
hidden field, and voila, they've edited the data.

If you have some way of revalidating the hidden data (if you require
HTTP_AUTHENTICATION, perhaps based on the REMOTE_USER) that would add a
level of security.  Better yet, use a cookie and store the data on your
server.  If the user can't access it, they can't modify it.

moo
----------------------------------------------------------------------
          Ron Klatchko - Manager, Advanced Technology Group           
           UCSF Library and Center for Knowledge Management           
                        ron at library.ucsf.edu



From holger at phoenix-edv.netzservice.de  Mon Aug  2 04:38:28 1999
From: holger at phoenix-edv.netzservice.de (Holger Jannsen)
Date: Mon, 2 Aug 1999 08:38:28 GMT
Subject: os.name
References: <1278785134-25350540@hypernet.com> <V5ho3.4340$VL2.463886@news.direcpc.com> <001e01beda8e$2111d2e0$f29b12c2@secret.pythonware.com> <Ggto3.4361$VL2.467372@news.direcpc.com> <7nuac5$3rs$1@chief.death.de> <RECo3.4365$VL2.468864@news.direcpc.com> <37A388DB.BFEB9A80@mail.matav.hu>
Message-ID: <37A55904.B3BAD2B8@phoenix-edv.netzservice.de>

Thank you,

that'll have to do it.

Ciao,
Holger

Arpad Kiss schrieb:
> 
> Emile van Sebille wrote:
> >
> > It sure would.  Any idea how you would get the result back into python
> > without having the dos command windows flash up on the screen?
> On Windows NT I use win32pipe:
> i,o,e=win32pipe.popen3('ver')
> ret=o.readlines()
> >
> > but-the-you-wouldn't-have-the-fun-of-parsing-the-file-ly yr's
> >
> > --
> >
> > Emile van Sebille
> > emile at fenx.com
> > -------------------
> >
> > Felix Puetsch <puetsch at gmx.de> wrote in message
> > news:7nuac5$3rs$1 at chief.death.de...
> > > emile at fenx.com (Emile van Sebille) wrote in
> > > <Ggto3.4361$VL2.467372 at news.direcpc.com>:
> > >
> > > >By the way, in windows at a command prompt:
> > > >
> > > >95,98:    net config
> > > >nt:       net config [server|workstation]
> > > >
> > > >gives a line that starts : Software version
> > >
> > > Wouldn't "ver" be easier?
> > >
> > > --
> > > Felix P?tsch
> >
> > --
> Arpad Kiss



From jpl at digex.net.nospamplease  Wed Aug 18 20:15:23 1999
From: jpl at digex.net.nospamplease (John Landahl)
Date: Wed, 18 Aug 1999 17:15:23 -0700
Subject: PIGgies?
Message-ID: <19990818171523.A22207@digex.net>

I'm a recent convert to Python and would love to see its credibility
heightened and its use in the mainstream increased.  Python seems to have
what it takes to become *the* general purpose programming language.  It's
certainly more suited for the job than some of the other current favorites.

Lately I've been considering forming a high profile user group in San
Francisco to help promote Python as a serious tool for mainstream use.  Much
to my chagrin I discovered that there's already a (recent) group called
"BayPIGgies".  I later noticed that the term "PIGgies" is used elsewhere for
Python user groups.  Am I the only Python enthusiast that has a problem with
the term "PIGgies"?

I certainly don't want to offend those who have chosen the term.  Rather, I
strongly believe that the Python community must remain united and close-knit
if the language is to gain wide-spread acceptance in the industry.  I think
that acceptance would be found more readily with advocacy groups bearing
slightly more serious names.  Of course, the "IG" in "PIGgies" stands for
Interest Group, so these groups are not necessarily meant to be public
advocates.  Even still, I think the term can still be a turn-off for those
thinking about getting involved with the language and the Python community.

These are my opinions, of course, and I graciously solicit the opinions of
others.  What do you think of the term "PIGgies"?  Do you feel that there is
a need (and room) for separate interest and advocacy groups?  Do either the
Python Software Activity or the Python Consortium already fill the advocacy
role?

John Landahl
jpl at digex.net



From clarence at silcom.com  Wed Aug 11 12:52:24 1999
From: clarence at silcom.com (Clarence Gardner)
Date: Wed, 11 Aug 1999 16:52:24 GMT
Subject: Tkinter newbie question
Message-ID: <cThs3.135$Vn5.17296@newsfeed.avtel.net>

So I've written my first GUI Python program (everything else I've done
for three years was CGI or automated stuff).  Very vanilla so far.  I'm
looking to do a little more, so I look at the Tk docs.  The question is,
how do you get from Tk to Tkinter.  For example, I want to place my top-
level window.  I see Tk_MoveToplevelWindow(), but I don't see how to
use it from Python.  Or maybe I can't.  (Hard to believe, but since I
see no call to it from _tkinter.c ???)

So, is there a general way to know what to do after you've found a Tk
function that you want to call?

Thanks.

-- 
-=-=-=-=-=-=-=-=
Clarence Gardner
AvTel Communications
Software Products and Services Division
clarence at avtel.com



From mstenber at cc.Helsinki.FI  Tue Aug  3 11:54:30 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 03 Aug 1999 18:54:30 +0300
Subject: Python microthreads
References: <FFropB.7tp@world.std.com> <933672146.479670@news.adtranzsig.de> <FFw8Bw.CI4@world.std.com>
Message-ID: <al87lncuafd.fsf@sirppi.helsinki.fi>

wware-nospam at world.std.com (Will Ware) writes:
> Dirk-Ulrich Heise (hei at adtranzsig.de) wrote:
> : Did you build it on top of Christian Tismers
> : Stackless Python, or is it a different kind of
> : modification to the VM?
> I haven't looked at Stackless Python, but it sounds like an unrelated
> thing. As I recall, Stackless Python uses the C stack for everything
> whereas normally Python has a piece of stack in each frame object.
> It can do this because the compiler is smart enough to figure out
> exactly how much stack will be needed for any function call. So
> frameobjects have a field called f_valuestack which maintains the
> arguments and local variables for that function.

No.. Stackless python does _not_ use C stack. Normal python does. AFAIK,
that's what the stackless stands for (purely frame objects and such).

> What I did was to keep the frameobject exactly as it is in normal
> Python, but invent a thread object with a stack for frameobjects,
> and a couple more stacks for stack pointers and program counters.
> Additionally, I modified the bytecode interpreter (eval_code2, in
> ceval.c) so that it runs for a specified number of opcodes and
> then stops, leaving the entire environment intact. This lets you
> rotate thru several threads, executing each for a few opcodes, so
> it looks like the threads are running simultaneously.

Sounds like something you could do with Stackless Python, or on top of,
fairly easily. 

-Markus

-- 

	"If you want to travel around the world and be invited to speak at
	a lot of different places, just write a Unix operating system."
		-- Linus Torvalds



From conversy at lri.fr  Mon Aug 30 09:17:00 1999
From: conversy at lri.fr (Stephane Conversy)
Date: Mon, 30 Aug 1999 15:17:00 +0200
Subject: Tkinter and window manager
Message-ID: <37CA844C.AE3F285@lri.fr>

Hi all.

I want the window manager to notify me when a window is raised
(typically
when I click on the bar...)

How to do this ?

    stef

--
St?phane Conversy
http://www-ihm.lri.fr/~conversy/
mailto:conversy at lri.fr






From pointal at lure.u-psud.fr  Fri Aug 27 04:07:20 1999
From: pointal at lure.u-psud.fr (Laurent POINTAL)
Date: Fri, 27 Aug 1999 08:07:20 GMT
Subject: Problem with sources - socketmodule.c
Message-ID: <37c643ca.348886171@news.u-psud.fr>

Hello,

I tried to recompile Python 1.5.2 on my home Linux box.
After a ./configure --with-threads, I tried a make... which fail on
the socketmodule.c.

Looking after the error, it was a problem of the number of arguments
for gethost_by_name_r (and for gethotstbyaddr_r). 
The ./configure has correctly tested my headers, found a 5 arguments
gethost_by_name_r, and defined HAVE_GETHOSTBYNAME_R5_ARG.

But in the socketmodule.c, the source begin by undefining the
preprocessor symbols HAVE_GETHOSTBYNAME_* set by ./configure, and set
them depending on the platform... and for linux it automatically set
it to HAVE_GETHOSTBYNAME_R6_ARG. Which is false on the Slackware4
Linux.

Currently, I have put a #if !defined(linux) / #endif around the
socketmodule.c code which modify HAVE_GETHOSTBYNAME_* and all works
well.

Where can I address that so the original source be modified to NOT
undefine the ./configure symbols in the case of a linux platform (I
dont know if other Unixes platform use ./configure too).

Thanks.

A+

Laurent.

Note:
Linux 2.2.6
libc.so.5.4.46
egcs 2.91.66
gcc 2.7.2.3
netdb.h v1.4 1995/08/14

---
Laurent POINTAL - CNRS/LURE - Service Informatique Experiences
Tel/fax: 01 64 46 82 80 / 01 64 46 41 48
email  : pointal at lure.u-psud.fr  ou  lpointal at planete.net 



From tim_one at email.msn.com  Thu Aug 26 21:22:35 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Thu, 26 Aug 1999 21:22:35 -0400
Subject: Fuzzy string matching?
In-Reply-To: <199908261746.TAA25440@dionysus.fw.cuci.nl>
Message-ID: <000901bef02a$a5296a00$35a2143f@tim>

[Hans Nowak]
> ...Anyone know of such a beast in Python?

Search for agrepy in comp.lang.python.announce, from 28 July 99.

all-questions-are-answered-anew-every-month<wink>-ly y'rs  - tim





From a.eyre at optichrome.com  Tue Aug 31 08:54:14 1999
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 31 Aug 1999 12:54:14 GMT
Subject: extension and core dumpage
In-Reply-To: <37C5DBC6.669869E7@tippett.com>
References: <37C5DBC6.669869E7@tippett.com>
Message-ID: <000801bef3af$edb04090$3acbd9c2@peridot.optichrome.com>

> SGI IRIX 6.3

Check out the notes in:

http://www.python.org/doc/1.5.1/ext/irixDynload.html

http://www.python.org/doc/1.5.1/ext/irixLinking.html

--------------------------------------------
Adrian Eyre <mailto:a.eyre at optichrome.com>
Optichrome Computer Solutions Ltd
Maybury Road, Woking, Surrey, GU21 5HX, UK
Tel: +44 1483 740 233  Fax: +44 1483 760 644
http://www.optichrome.com 
--------------------------------------------




From nathan at islanddata.com  Thu Aug 19 18:00:18 1999
From: nathan at islanddata.com (Nathan Clegg)
Date: Thu, 19 Aug 1999 15:00:18 -0700 (PDT)
Subject: exception table
Message-ID: <XFMail.990819150018.nathan@islanddata.com>

Has anyone yet compiled a table of common functions (both built-in and
imported) and all possible exceptions?  It seems to me this would help in
bullet-proofing applications as a quick reference check on whether or not
something was missed.  Any pointers to such a thing?


----------------------------------
Nathan Clegg
 nathan at islanddata.com





From srenner at lycosmail.com  Thu Aug 26 20:05:43 1999
From: srenner at lycosmail.com (srenner at lycosmail.com)
Date: Fri, 27 Aug 1999 00:05:43 GMT
Subject: I can't open a named FIFO that already exists.
Message-ID: <7q4koc$h0u$1@nnrp1.deja.com>

I have an object in my Linux filesystem that I created with 'mkfifo'. I
suppose it's called a FIFO. Anyway, the idea is to use it as a pipe. It
already exists. I want to write lines to this pipe so that another
program (Csound) can be controlled in real time by reading events from
it. But this object can't be opened inside Python, because neither
os.popen('fifoname') nor open('fifoname') returns.

I don't want to open a pipe inside Python, because I have to provide a
device name to Csound on its command line, and this way it already has a
name.

I'm sure I'm missing something. Any help would be appreciated. sr


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mwh21 at cam.ac.uk  Sat Aug 14 07:20:25 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 14 Aug 1999 12:20:25 +0100
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk> <luemh851ym.fsf@stest.ch> <YeMs3.6620$gO1.175350@news2.giganews.com> <m3u2q42ole.fsf@atrus.jesus.cam.ac.uk> <2z3t3.7359$gO1.213382@news2.giganews.com>
Message-ID: <m3r9l638xy.fsf@atrus.jesus.cam.ac.uk>

cbbrowne at news.hex.net (Christopher Browne) writes:

> On 13 Aug 1999 07:15:25 +0100, Michael Hudson <mwh21 at cam.ac.uk> wrote:
> >cbbrowne at news.hex.net (Christopher Browne) writes:

[snip]

> >> Sounds Really A Lot like CORBA DII...
> >
> >Not to me. Could you elaborate for the slow of thought?
> 
> DII == Dynamic Invocation Interface.
> 
> Basic idea is that it allows a CORBA client to dynamically discover
> new objects and their interfaces, and connect to them.
> 
> Same idea as having a LibGlade client dynamically discover new GTK
> widgets and their interfaces, and then connecting to them.
> 

Oh, I see. 

I'd hypothesise that this more the result of unfortunate overloading
of the word "interface" than anything fundamentally similar. 

But it's still pretty cool, particularly when you're using a dynamic
language like Python.

Regards,
Michael



From watterss at pilot.msu.edu  Sat Aug 21 21:35:46 1999
From: watterss at pilot.msu.edu (Shannon Watters)
Date: Sat, 21 Aug 1999 21:35:46 -0400
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu>
Message-ID: <7pnk99$35d$1@msunews.cl.msu.edu>

Tom Christiansen <tchrist at mox.perl.com> wrote in message
news:37bd5e2a at cs.colorado.edu...
>
>  *  I want to be able to talk to other people in their own language.
>  *  I want to read their literature.
>  *  I want insights into my own tongue.
>

No, you just want to be a pain in the ass.  You're also extremely arrogant.
This is obvious.





From tchrist at mox.perl.com  Thu Aug 19 11:33:38 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 08:33:38 -0700
Subject: Kudos to documentors: (popen2)
References: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de>
Message-ID: <37bc15c2@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Robb Shecter <shecter at darmstadt.gmd.de> writes:
:I needed to do some system programming, and at first tried to use
:Perl.  Although I've used Perl for years, this documentation page
:forced me to give Python a try for it:
:
:http://www.perl.com/CPAN/doc/manual/html/lib/IPC/Open2.html

The standard documentation for Perl is found in manpages, not
on the web.  That's an old copy.

    man IPC::Open2

would have answered it.

:(Just notice how much is implicit, and not documented, and with no
:links to other needed info.)

I have no idea what you're talking about, or what links you're
looking for, but as the original author of that module, I am 
always happy to accept documentation suggestions.

Or is this just another case of pissing on something that you didn't
understand out of the way of the people responsible?  If so, you lose,
because *I'm* responsible.  What's your biggest problem?

And remember, Perl's documentation is NOT WEB-BASED.  That's 
merely an automatic conversion, and suffers various faults
because of that.  It is not maintained in that form, nor does
it cater to it.

Admittedly the docs for open2 there in the module are very sparse.
Did you read the section on "Bidirectional Communication with Another
Process" in the standard perlipc manpage, which is the standard tutorial
for doing interprocess communication in Perl?  Here's a relevant excerpt:

       Bidirectional Communication with Another Process

       While this works reasonably well for unidirectional communication,
       what about bidirectional communication?  The obvious thing you'd
       like to do doesn't actually work:

           open(PROG_FOR_READING_AND_WRITING, "| some program |")

       and if you forget to use the -w flag, then you'll miss out entirely
       on the diagnostic message:

           Can't do bidirectional pipe at -e line 1.

       If you really want to, you can use the standard open2() library
       function to catch both ends.  There's also an open3() for
       tridirectional I/O so you can also catch your child's STDERR,
       but doing so would then require an awkward select() loop and
       wouldn't allow you to use normal Perl input operations.

       If you look at its source, you'll see that open2() uses low-level
       primitives like Unix pipe() and exec() calls to create all the
       connections.  While it might have been slightly more efficient by
       using socketpair(), it would have then been even less portable than
       it already is.  The open2() and open3() functions are  unlikely to
       work anywhere except on a Unix system or some other one purporting
       to be POSIX compliant.

       Here's an example of using open2():

           use FileHandle;
           use IPC::Open2;
           $pid = open2(*Reader, *Writer, "cat -u -n" );
           Writer->autoflush(); # default here, actually
           print Writer "stuff\n";
           $got = <Reader>;

       The problem with this is that Unix buffering is really going to
       ruin your day.  Even though your Writer filehandle is auto-flushed,
       and the process on the other end will get your data in a timely
       manner, you can't usually do anything to force it to give
       it back to you in a similarly quick fashion.  In this case,
       we could, because we gave cat a -u flag to make it unbuffered.
       But very few Unix commands are designed to operate over pipes,
       so this seldom works unless you yourself wrote the program on
       the other end of the double-ended pipe.

	[verba deleta]

The section following that one is on "Bidirectional Communication with
Yourself", which discusses Perl's forking open call.  You can also
learn about this in the recently added perlopentut manpage, which is
a tutorial on how to open various things.  Sounds like you could use a
bit of a tutorial.

Next time, please try the manpages, and send the authors mail or file a
bug report.  But please don't just piss on people where you think they
can't see you.  Sheesh.

--tom
-- 
Those who do not fight the demons within themselves are destined to fight them 
 outside themselves...



From Ted.Horst at wdr.com  Mon Aug 30 14:07:41 1999
From: Ted.Horst at wdr.com (Ted Horst)
Date: Mon, 30 Aug 99 13:07:41 -0500
Subject: 'if foo' calls getattr?!?
In-Reply-To: <joe-300819991050234291@chinacat.salk.edu>
References: <joe-300819991050234291@chinacat.salk.edu>
Message-ID: <199908301807.AA02916@ch1d2833nwk>

On Mon, 30 Aug 1999, Joe Strout wrote:
> Doing some profiling on Graphite this weekend, we found a very curious
> thing: a gazillion calls to __getattr__ on one of our classes, where we
> expected to find none.  We eventually tracked it down to a seemingly
> innocent test, basically, "if foo: self.myfoo = foo".  The "if foo" was
> invoking foo.__getattr__ (which is very expensive).
>
> I can think of no reason why it should be this way, but it is.  Below is
> a very simple stand-alone demonstration.  It leads me to the absurd
> idiom of saying "if id(foo) != id(None)" when I really mean "if foo",
> just because using id() avoids the __getattr__ call.
>
> Can anybody explain why this should be?  Is it a bug in Python?  Is
> there some other way to check whether I have a reference to an object
> that doesn't involve a function call?
>
> Confused,
> -- Joe

The statement

if foo:

tries to determine the truth value of foo.  Different types respond to this  
differently.  An instance will respond by returning the results of  
__nonzero__ if defined, or else returning the result of __len__ if defined.

The following session illustrates this, and a faster way to check for None.

>>> class Foo:
...     def __getattr__(self,name):
...             print name
...             raise AttributeError, name
...
>>> foo = Foo()
>>> if foo:
...     print 'yes'
...
__nonzero__
__len__
yes
>>> if foo is not None:
...     print 'yes'
...
yes
>>>

Ted

---



From philh at vision25.demon.co.uk  Tue Aug 24 15:34:35 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 24 Aug 99 19:34:35 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <37c2a462.349531009@192.77.87.74>
Message-ID: <935523275snz@vision25.demon.co.uk>

In article <37c2a462.349531009 at 192.77.87.74> jlj at cfdrc.com "Lyle Johnson" writes:
> >
> >This is tcl/tk. Almost no difference. A bit more verbose, but there are
> >numerous ways to shorten code, which I haven't used intentionally until
> >I got tired and used a loop to produce four radiobuttons.
> >
> 
> I agree that you're not saving yourself a lot of typing if you're only
> going to do a Tk implementation of your GUI, or only a GTK
> implementation, etc. But the point of this guy's "Parrot" is that you
> can plug in different back-ends to generate, e.g. Tcl scripts or HTML
> forms or C/GTK code or whatever.
> 
> I *personally* don't have a need to generate different implementations
> of my GUIs in this way; but if I did I could see this tool being
> useful.


The Parrot tool has other potential advantages, e.g: 

1. you only have to learn one syntax, that can then be used for
lots of languages and GUI libraries.

2. assuming the right import and export tools have been written,
Parrot can allow you to use one GUI builder to output to a different
GUI, for example you could design your GUI in Glade, and use Parrot
to output it as a series of C++ files for KDE/Qt.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From mlh at idt.ntnu.no  Sun Aug  8 09:42:03 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 08 Aug 1999 15:42:03 +0200
Subject: lambda fun: infinite streams
References: <37ABADB1.834A664E@go2net.com> <slrn7qnsod.jjr.thantos@brimstone.mecha>
Message-ID: <y0jlnbmgzis.fsf@drue.idi.ntnu.no>

thantos at brimstone.mecha (Alexander Williams) writes:

> On Fri, 06 Aug 1999 20:53:21 -0700, Joe Bowbeer <joeb at go2net.com> wrote:
[...]
> Its good to see other Scheme/Fnctional-language freaks hanging around
> with the Python folks, especially those with copies of SICP.
> 
> Now, if only Python had lazy evaluation and Streams as first-class
> entities, my life would be complete.

Wow... ;)

> My life would be decently
> complete anyway if map() worked on your infinite streams as stands.
> ;)
> 

Well... You could make one yourself...

<treading carefully, as I am not really a functional programmer...
Dropping the multiple stream version for now:>


def mapStream(f,s):
    return (apply(f,[head(s)]),
           lambda f=f, s=s: mapStream(f,tail(s)))


--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From skip at mojam.com  Fri Aug  6 13:07:08 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri,  6 Aug 1999 12:07:08 -0500 (CDT)
Subject: Proper application of the buffer interface
Message-ID: <14251.4680.760818.979812@dolphin.mojam.com>

Sorry to be a numbskull, but since there is apparently little or no
documentation for the buffer interface yet, can someone give me a thumbnail
sketch of why we have it and what it's used for?  I can guess that there are
some situations where using string objects as raw buffers could lead to
misinterpretation of their contents, especially since not everything is
US-ASCII anymore.  Beyond that, I'm not sure why there would be a
distinction between buffers and strings.

Thx,

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098




From hnowak at cuci.nl  Tue Aug 31 17:35:03 1999
From: hnowak at cuci.nl (Hans Nowak)
Date: Tue, 31 Aug 1999 21:35:03 GMT
Subject: tkinter/python question
In-Reply-To: <014e01bef2bd$3f8edaa0$f29b12c2@secret.pythonware.com>
References: <014e01bef2bd$3f8edaa0$f29b12c2@secret.pythonware.com>
Message-ID: <199908312136.XAA25476@dionysus.fw.cuci.nl>

On 30 Aug 99, at 9:53, Fredrik Lundh wrote:

> and there are (at least) two Tkinter books in the
> works; one by John Grayson (for Manning) and one
> by yours truly (for O'Reilly).

When will the book come out, Fredrik?

--Hans Nowak (zephyrfalcon at hvision.nl)
Homepage: http://fly.to/zephyrfalcon
Python Snippets: http://www.hvision.nl/~ivnowa/snippets/
The Purple Kookaburra Forum: http://www.delphi.com/kookaburra/



From gmcm at hypernet.com  Fri Aug 27 21:11:38 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 27 Aug 1999 20:11:38 -0500
Subject: Why use Python when we've got Perl?
In-Reply-To: <37C60262.C6771B75@megsinet.net>
Message-ID: <1276323665-38423440@hypernet.com>

> Where can I find info on Minotaur?
> 
> David O.

http://mini.net/pub/ts2/minotaur.html

It's really proof of concept, (don't expect too much). There also
hasn't been much activity there in a month or so. I don't think it's
dead, but I think it will be awhile before the main actors get back
to it. 

- Gordon



From moun at usenix.org  Mon Aug 16 19:21:23 1999
From: moun at usenix.org (Moun Chau)
Date: Mon, 16 Aug 1999 23:21:23 GMT
Subject: USENIX Tcl/Tk Conference Call for Papers-Paper submissions due soon: 9/1/99
Message-ID: <FGKzJn.LwF@usenix.org>

The 7th USENIX Tcl/Tk Conference is a forum to:
 * bring together Tcl/Tk researchers and practition
 * publish and present current work involving Tcl/Tk
 * learn about the latest developments in Tcl/Tk
 * plan for future Tcl/Tk related developments

Tcl/2K : The 7th USENIX Tcl/Tk Conference
      February 14-18, 2000
      Austin, Texas, USA
Sponsored by USENIX, The Advanced Computing Systems Association
---------------------------------------------------------------------------- 
Please find the Call for Papers at
http://www.usenix.org/events/tcl2k/cfp
----------------------------------------------------------------------------
Paper submissions due: September 1, 1999
Demonstration, and Panel Proposals due: September 1, 1999
Poster submissions due: December 8, 1999

The conference program will include formal paper and panel presentations,
poster and both reviewed and informal demonstrations, works-in-progress,
Birds-of-a-Feather sessions, and two-days of high-quality tutorials.  All
forms of participation provide an opportunity to report on original Tcl/Tk
research.

Formal papers should address topics of interest to experienced Tcl/Tk
programmers; posters and informal demos may be geared to any level of user
from beginner to expert. Awards will be given for the best
paper and best student paper at the conference.
============================================================================
The USENIX Association's international membership includes engineers,
system administrators, and computer scientists working on the cutting edge
of  systems and software.  Our conferences are recognized for their
technical excellence and pragmatic emphasis.






From tchrist at mox.perl.com  Thu Aug 19 13:08:32 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 10:08:32 -0700
Subject: Kudos to documentors: (popen2)
References: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de> <37bc15c2@cs.colorado.edu> <37BC21D9.E19DBEA2@darmstadt.gmd.de>
Message-ID: <37bc2c00@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Robb Shecter <shecter at darmstadt.gmd.de> writes:
:It ought to be.  Or at least, well-linked "info" pages.  (This is
:surely easily agreed upon, right?)  

You'd be surprised.

But this is a moot issue, because:
:>     man IPC::Open2
:...has apparently the same text that bothered me.  
:
:
:1)  It will not create these file handles for you.  You have
:       to do this yourself.  So don't pass it empty variables
:       expecting them to get filled in for you. 

:
:My question: how do I create these file handles?  What document would
:I look in to find out?  

Normally you just use them, as in

    open2(FH1, FH2, "cmd here")

and they appear as globals in your package namespace (that's module
namespace in python parlance).  People don't usually create anon handles.
They can be localized to a scope, though, as in this excerpt from the
perldata manpage:

       sub newopen {
	   my $path = shift;
	   local *FH;    # create new filehndle
	   open   (FH, $path) or  return undef;
	   return *FH;
       }
       $fh = newopen('/etc/passwd');

Which might suffice for your purposes.  Filehandles are not very nice in
Perl as some level.  There are object-oriented modules like IO::Handle
and IO::File, but these are just syntactic saccarine that come at a cost
of 10-100x normal built-in operations, and so they're eschewed by all
except that OO zealouts.  They do let you do

    use IO::Handle;
    $rdr = IO::Handle->new();
    $wtr = IO::Handle->new();

Which is less mysterious than the cheap and fast way:

    $rdr = do { local *FH };
    $wtr = do { local *FH };

You're correct that the documentation for the module is very sparse,
and assumes that you know this.

--tom
-- 
    "If ease of use is the highest goal, we should all be driving golf carts."
    	--Larry Wall



From jturner at ug.cs.su.oz.au  Mon Aug 30 18:38:20 1999
From: jturner at ug.cs.su.oz.au (Jeff turner)
Date: Tue, 31 Aug 1999 08:38:20 +1000
Subject: Question: lists in classes
References: <37CA1803.2FA9261B@ug.cs.su.oz.au>
Message-ID: <37CB07DC.A3AAB228@ug.cs.su.oz.au>

Thanks to everyone who replied. It -was- in the FAQ (6.25).. I'm
suitably shamed ;)




From wish at dumain.demon.co.uk  Tue Aug 31 07:42:14 1999
From: wish at dumain.demon.co.uk (Bill Hay)
Date: 31 Aug 1999 11:42:14 GMT
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net> <99082819072100.03655@o112> <7q98g5$7bi$5@newshost.accu.uu.nl>
Message-ID: <slrn7snfsl.foo.wish@dumain.com>

In article <7q98g5$7bi$5 at newshost.accu.uu.nl>, Martijn Faassen wrote:
>Oh my god they KILLED Richard Stallman!!
>
You Bastards!

-- 
TANSTAAFM



From kosullivan at snet.wit.ie  Tue Aug 24 08:07:13 1999
From: kosullivan at snet.wit.ie (Kieran O' Sullivan)
Date: Tue, 24 Aug 1999 13:07:13 +0100 (IST)
Subject: Can't use crypt
Message-ID: <Pine.LNX.4.10.9908241244050.21229-100000@snet.wit.ie>

When I tyr to compile python 1.5.2 on SuSE Linux 6.0 with the crypt module
I get the following error.

make[1]: Entering directory `/Python-1.5.2/Modules'
gcc  -Xlinker -export-dynamic python.o \
          ../libpython1.5.a   -lieee -ldl  -lm  -o python
../libpython1.5.a(cryptmodule.o): In function `crypt_crypt':
/Python-1.5.2/Modules/./cryptmodule.c:21: undefined reference to `crypt'
collect2: ld returned 1 exit status
make[1]: *** [link] Error 1
make[1]: Leaving directory `/Python-1.5.2/Modules'
make: *** [python] Error 2

I have compiled python 1.5.2 with crypt on SuSE 5.2 with no problems.
There was a previous copy of python 1.5.1 on the SuSE 6.0 machine but
python 1.5.2 cpmpiles without any problem when crypt is not included.

When I use -lcrypt instead python will compile but I can't use cthe crypt
function.





From evan at tokenexchange.com  Thu Aug 12 17:34:10 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Thu, 12 Aug 1999 16:34:10 -0500
Subject: Help!  $PYTHONPATH trouble
References: <37B33980.A5318181@n-space.org>
Message-ID: <n5Hs3.14$2y4.6929@news-dal.corridex.com>

First, you need to either export PYTHONPATH or use the assignment as a
prefix, thus (assuming bash):

$ PYTHONPATH="home/possum/test/kwp"
$ export PYTHONPATH
$ python
or
$ PYTHONPATH="home/possum/test/kwp" python

Second, shouldn't that path start with a "/"?

Erik Stewart <possum at n-space.org> wrote in message
news:37B33980.A5318181 at n-space.org...
The variable $PYTHONPATH doesn't seem to have any effect on my system (RH
Linux 6.0, Python 1.5.1).  Here's what I get after defining $PYTHONPATH:






From bwarsaw at cnri.reston.va.us  Wed Aug 11 15:48:09 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Wed, 11 Aug 1999 15:48:09 -0400 (EDT)
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au>
	<m3vham44g5.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <14257.54137.72005.705719@anthem.cnri.reston.va.us>

>>>>> "MH" == Michael Hudson <mwh21 at cam.ac.uk> writes:

    MH> I think JPython does some global analysis; not heard anything
    MH> on that front for a while though.

At IPC7, JimH reported on some experiments he'd been doing
<http://www.jpython.org/jpython-talk-1.ppt> with whole program
analysis in the jpythonc tool.  He reported some impressive results,
but alas most of that stuff had to be taken out of the released
version of the tool.  Due to lack of time, this work will probably be
on the back burner for a long while unless someone else with interest
adopts it.

-Barry



From schorsch at schorsch.com  Sun Aug 22 11:29:49 1999
From: schorsch at schorsch.com (Georg Mischler)
Date: Sun, 22 Aug 1999 15:29:49 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk>
Message-ID: <7pp517$dce$1@nnrp1.deja.com>

philh at vision25.demon.co.uk wrote:
> It looks like version 0.0.1 of Parrot, my text-based GUI builder,
> will be ready within a day or so. It doesn't actually do anything
> useful yet; v0.0.1 is a technology demonstration version, which
> is intended to verify that the basic design is more-or-less
> sensible. Would anyone like to have a look at it, and possibly
> suggest some improvements? If so, tell me, and I'll put it on
> my website.
[...]
> To use Parrot, first you create a *.par file describing your
> application's GUI. Your file might look something like this:
>
> -----------Cut here-------------------------------
> window @MyWindow "My First Window" {
>    menuBar {
>       menu "File" {
>          menuItem @New "New"
>          menuItem @Exit "Exit"
>       }
>       menu "Help" {
>          menuItem @About "About..."
[...]


Looks very interesting.
And actually reminds me a lot of autodesks slowly deprecated
dcl description language which I also liked a lot at the time.
Is this a coincidence?

Another thing: Do you have any concepts on supporting dynamic
dialogs with your tool? The most obvious method of course being
able to enable and disable widgets on the fly, but it would
also be nice to show and hide them while the window is displayed.
I don't assume this would be trivial to implement, especially
at such an early stage in the developpement. Maybe leaving
the possibility open to "include" and "remove" subsets of a
description during runtime would do the trick, but you probably
have more insight in the potential difficulties than I do.


Have fun!

-schorsch

--
Georg Mischler  --  simulation developper  --  schorsch at schorsch.com
+schorsch.com+  --  lighting design tools  --  http://www.schorsch.com/


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From ralf.doering at prakinf.tu-ilmenau.de  Tue Aug  3 02:33:00 1999
From: ralf.doering at prakinf.tu-ilmenau.de (Ralf Doering)
Date: 03 Aug 1999 08:33:00 +0200
Subject: Annoying problem with rexec
Message-ID: <xzn1w99xwj.fsf@prakinf.tu-ilmenau.de>

Hello,
maybe someone can explain what happens here. Like stated in the
subject, the problem is about the rexec module. I tracked it down to
this simple example:

def testrexec():
    ml = ['test.1','test.2','test.3']
    for mod in ml:
        temp = "/tmp/"+mod
        print "**** About to open:", temp
        f2 = open(temp,"w")
        f2.write("I WAS HERE.\n")
        f2.close()
        print "****",temp, "closed ."
        saved_open = open

        import rexec
        r = rexec.RExec(verbose=1)
        open = r.r_open
        #do some stuff here like load_modules ...
        pass
        open = saved_open

As a result, I get:

>>> testrexec.testrexec_ok()
**** About to open: /tmp/test.1
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "testrexec.py", line 8, in testrexec
    f2 = open(temp,"w")
NameError: open
>>> 


I do not understand, why open becomes undefined, even *before* rexec
is imported and before I try to do nasty things with the function
reference. I did expect, that the first call to "open" in write mode
should succeed, even if something goes wrong with the references to
open later. But it seems, that I`m  wrong.
However, if I insert a "global open" right after the start of the
function, it works as expected.
Anyone out there how can enlight me?

Ralf
-- 
No sig -- no fun



From tismer at appliedbiometrics.com  Thu Aug 19 11:12:20 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Thu, 19 Aug 1999 15:12:20 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <37BC1ED4.A29883BC@appliedbiometrics.com>


Tom Christiansen wrote:

[snipped a long list which I commented via private emails already]

> GOTCHA: (low)
>     Python has a round() function, but it seems to ignore the
>     IEEE rules of rounding.  However, its sprintf operator doesn't:
>         >>> print "int of %f is %.0f" % (1.5, 1.5)
>         int of 1.500000 is 2
>         >>> print round(1.5,0)
>         2.0
>         >>> print "int of %f is %.0f" % (2.5, 2.5)
>         int of 2.500000 is 2
>         >>> print round(2.5,0)
>         3.0

I had a look at this again. Despite the fact that the text in
the above example is a bit misleading by "int of" where
a round() takes place, which version of Python/architecture
did you use? I get the expected result, see below.

>>> print "round of %f is %.0f" % (1.5, 1.5) 
round of 1.500000 is 2
>>> print round(1.5,0)
2.0
>>> print "round of %f is %.0f" % (2.5, 2.5)
round of 2.500000 is 3
>>> print round(2.5,0)
3.0
>>> 

>     And I'd jolly well like to know why I wasn't allowed to use
>         print "int of %f is %.0f" % (2.5) * 2
>     or if needed,
>         print "int of %f is %.0f" % ( (2.5) * 2 )

This is just a matter of commas needed for singleton tuples,
and operator precedence, yielding

>>> print "round of %f is %.0f" % ((2.5,) * 2)
round of 2.500000 is 3
>>> 

ciao - chris

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From urner at alumni.princeton.edu  Mon Aug 23 16:16:26 1999
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Mon, 23 Aug 1999 20:16:26 GMT
Subject: Pro Python remarks to math ed folks (pointers)
Message-ID: <37c2a51e.18050990@news.teleport.com>


Some "Python as teaching language" emphasis in my recent
posts to elsewhere.  For those of you wanting to see how
I'm using Python with polyhedra, check out:

http://forum.swarthmore.edu/epigone/geometry-research/whelsendsnimp
http://www.deja.com/getdoc.xp?AN=516251867&fmt=text

My basic approach is to hit two key topics in tandem in
K-12:  computer programming and polyhedral geometry:  by
making polyhedra my paradigm objects (in the OOP sense,
but also in a literal easy-to-understand sense).

I'll add here (given this ng is for Python speakers) how 
happy I am with operator overloading, which I haven't 
had before, given my own path through the languages has
so far mostly skirted C/C++.

With a polyhedron object in P (say an icosahedron), Q=P*3 
is another polyhedron (same shape), but 3x larger, and 
the equally simple expression R=Q+v creates polyhedron R 
translated through space by vector v.  Short, sweet, to
the point.

The fact that a class is built around built-in __dict__ 
made it easy to go forward with concepts I'd already used
in the Java version of this exercise.  I have points A-Z 
as a set of reference points for anchoring a smallish set 
of polyhedra.  

In Java, I added lookup strings 'A'-'Z' with their associated 
vectors to a Hashtable.  In Python, a dictionary is even 
easier to use (same idea as Hashtable of course), and, even 
better, such a dictionary is already part of any class.  
Just define A=Vector(xyz=(1,0,0)) and you've got your 
{'A':Object} linkup for free!  Way cool.

Kirby
Curriculum writer
Oregon Curriculum Network
Cc: Synergetics-L at telelists.com

PS:  I've posted to this newsgroup once before so far:
http://www.deja.com/getdoc.xp?AN=490567169&fmt=text





From pehr at alum.mit.edu  Fri Aug  6 00:03:40 1999
From: pehr at alum.mit.edu (pehr anderson)
Date: Fri, 06 Aug 1999 00:03:40 -0400
Subject: Serial port IO question
References: <379DF55E.EAFCBBF7@hti.com>
Message-ID: <37AA5E9C.678580E4@alum.mit.edu>

Dear David,
Don't forget, you have to set all the serial port parameters correctly.
The easiest way to do this is with stty. For example, under linux add the
line:
    os.system("stty 19200 raw < " + upsdev)
You need to find the appropriate speed of course...
For more information see "man stty"... it's not very informative but at
least it works.
    -pehr


David Stokes wrote:

> I am attempting to send a ^M (Control-M) down a serial port and then
> read back from that same port.  The application is an old UPS that will
> send back system status after a ^M.  The code below opens the port but
> the readline seems to be returning what I wrote.  My guess is that I
> haven't really grabbed I/O stream for the read operation.  Any
> suggestions?
>
> Dave Stokes
> -------
>
> #!/usr/bin/env python
> import sys, string
>
> upsdev = "/dev/ttyS1"   # Serial port attached to ups
>
> class   get_ups:
>     def __init__(self):
>         print "Getting ups status"
>
>     def get_ups_data_p(self):
>         upsdata = open( upsdev, 'r+');      # open port to ups
>         upsdata.write("\015")                   # send ^M to ask for
> status
>         upsdata.flush()
>         upsbuffer = upsdata.readline()       # get status back
>         upsdata.close()
>
> if __name__ == '__main__':
>     ups_status = get_ups()
>     ups_status.get_ups_data_p()




From marc.saric at mpi-dortmund.mpg.de  Wed Aug 25 11:06:22 1999
From: marc.saric at mpi-dortmund.mpg.de (Marc Saric)
Date: Wed, 25 Aug 1999 17:06:22 +0200
Subject: Which GUI-toolkit
Message-ID: <37C4066E.9EB52B39@mpi-dortmund.mpg.de>

Hi All

Just a newbie question: I want to write a GUI-tool for a
FORTRAN-program-package (communication goes via input-and configuration
files for that program for now) and I had a look at wxWindows and
Tkinter. 
To me it seems that wxWindows provides a lot more prebuild highlevel
parts like notebooks and tree views. On the other hand Tk might be more
widespread, especially on UNIX (I plan to do the program for both UNIX
and Windows with a focus on UNIX due to the FORTRAN package).

Any advice to which GUI I should use??

-- 
Bye,

Marc Saric

Max-Planck-Institut f?r molekulare Physiologie
Otto-Hahn-Strasse 11
44227 Dortmund

phone: +49 231 133 2168



From esuzm at dcs.warwick.ac.uk  Fri Aug  6 13:27:15 1999
From: esuzm at dcs.warwick.ac.uk (Andrew J Clover)
Date: 6 Aug 1999 18:27:15 +0100
Subject: Win: ADO and Index Server through Python
Message-ID: <7of5tj$7km$1@stone.dcs.warwick.ac.uk>

Hmm. I'm not convinced this is a Python problem, but there doesn't seem to
be an appropriate group for Index Server, unless I've missed something, and
people using VB in ASPs seem to be able to get this stuff to work, so...

I'm using some Python CGIs to access MS Index Server through ADO using SQL
queries. Python and Index Server *seem* to be talking to each other, but
Index Server never returns *any* results no matter what the query is. The
index-building process seems to work fine; MMC reports I've got 1Mb of
catalog information and 34 documents indexed.

A test script follows. Does anyone with experience of using Index Server from
Python have any idea why Index Server might always return "0 results"? Am I
missing the point and doing something stupid? (Wouldn't be the first time.)

Or does anyone have any recommendations of a good non-Win-specific search
engine I could tie in instead of this nasty MS stuff?

# testIndexServer.py

import win32com.client
com= win32com.client.Dispatch
con= win32com.client.constants

query= "\
SELECT Filename, Path, DocTitle \
FROM SCOPE() \
WHERE CONTAINS (Contents, 'law')>0"

# any other query can be used with similar results; even a query with no
# 'WHERE' clause returns zero results

connection= com('ADODB.Connection')
connection.Provider= 'MSIDXS'
connection.Open('GlobalSite')
result= com('ADODB.RecordSet')
result.MaxRecords= 200
result.Open(query, connection, con.adOpenStatic)

print str(result.RecordCount)+' results'
while not (result.EOF==-1):
  print result.Fields('Filename')
  result.MoveNext()

connection.Close()

# help me

-- 
This posting was brought to you by And Clover.
(Sorry.)



From frankn=news at cs.vu.nl  Mon Aug 23 03:37:56 1999
From: frankn=news at cs.vu.nl (Frank Niessink)
Date: Mon, 23 Aug 1999 07:37:56 GMT
Subject: Tkinter and classes
References: <37C0349A.EF5EEA6D@uniserve.com>
Message-ID: <7pqtok$lp3@cs.vu.nl>

Justin Muir <jkmuir at uniserve.com> wrote:

> class Helloworld:
>     def __init__(self, master):
>             _lblstr = StringVar()
>             _lblstr.set("hello world")

>             frm = Frame(master)
>             frm.pack()

>             but = Button(frm, text='Hello!', command=setLabel)
>             but.grid(row=0)

>             lbl = Label(frm, textvariable=_lblstr)
>             lbl.grid(row=1)

>     def setLabel(self):
>             _lblstr.set("HELLO WORLD!!")

> This will run, but barfs out on the callback.
> Am I misunderstanding the relationship between the callback to the
> namespace of the class?

Try replacing _lblstr by self._lblstr to make it an instance variable.

As you programmed it above _lblstr is just local to the __init__ function
and disappears as soon as the function is finished. So setLabel doesn't 
know anything about a _lblstr.

Cheers, Frank

-- 
"I think we have different value systems."  "Well mine's better."  "That's
according to your ... oh never mind."
	-- Douglas Adams, 'Mostly Harmless'



From greg.ewing at compaq.com  Tue Aug 31 00:24:18 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Tue, 31 Aug 1999 16:24:18 +1200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <yaZu3.23$2y4.21931@news-dal.corridex.com> <37C1B864.970C63DC@compaq.com> <b2yw3.31$2y4.39902@news-dal.corridex.com>
Message-ID: <37CB58F2.A88075CF@compaq.com>

Evan Simpson wrote:
> 
> Sweet!  Are you doing dicts too?  I await the day when I can litter my code
> with calls like "fn(1, *twople, 'with a melon?', *threeple, **websters,
> **randomhouse)".

Currently you're allowed at most one tuple and one dict,
and they have to be at the end of the parameter list,
by analogy with the def syntax.

This seems to be adequate for the main intended use --
passing parameters on to inherited methods. Doing
anything more complicated would be rather awkward
to implement.

Greg



From skaller at maxtal.com.au  Tue Aug 10 18:37:58 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Tue, 10 Aug 1999 22:37:58 GMT
Subject: Q: When and why to use dynamic variables?
References: <37AED975.2DDF2410@darmstadt.gmd.de>
Message-ID: <37b0a506.2379413816@news.triode.net.au>

On Mon, 09 Aug 1999 15:36:53 +0200, Robb Shecter <shecter at darmstadt.gmd.de> wrote:

>Hi,
>
>I'm looking for more information about the ability in Python to add
>variables to an object instance at runtime.  I'm interested in both
>the theoretical as well as practical implications of this feature.
>
>Can anyone point me to information about it?  For example, code where
>this feature is used, descriptions of it, suggested uses, etc.

	Actually, you have it backwards: Python cannot do
anything else. New class instances never have _any_ attributes,
nor do new modules. They have to be added dynamically,
in the class case by executing the __init__ routine,
or in the module case by executing the module.

	So the reverse question is much more important:
how can we compile optimised code if everything is so dynamic?

	There is an answer for functions, that many of the variables
are known, and the compiler generates special instructions referring
to them by number, and storing them in an array.

	The 'theoretical' implications of dynamic attributes
are that optimisation and reasoning about your code is
not easy.

	Here is a 'practical' use of the feature.
I have a program called interscript that weaves special
files into documents. It can weave TeX, HTML, plain
text, and other formats from the same sources.
These weavers are represented by classes
for each kind of document format.

	Now, sometimes, one needs to provide
different data to different kinds of weaver,
but the control program only knows the _abstract_
nature of a weaver. So it makes sense to pass
a whole set of 'arbitrary' attributes to weavers,
and each weaver can use the data if it applies,
or ignore it otherwise.

	Similarly, in order to allow clients
to provide arbitrary arguments on the command
line, which can be accessed inside a document,
I can just use dynamic attributes.

	In both these cases I'm implementing
a kind of 'miniinterpreter' in which I leverage
Pythons own interpreter to do most of the work.
Indeed, interscript's 'mini' interpreter is
full blown python plus more.

	Here's some code:

	# file x.py
	import sys
	def p(x): print x
	exec sys.argv[1]

now,

	python x.py "x=1; y=2; p(x+y)"

which demonstrates the idea.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From philh at vision25.demon.co.uk  Sun Aug  8 17:30:16 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 08 Aug 99 21:30:16 GMT
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk> <N8ir3.937$zN2.21156@typhoon.nyroc.rr.com>
Message-ID: <934147816snz@vision25.demon.co.uk>

In article <N8ir3.937$zN2.21156 at typhoon.nyroc.rr.com>
           news at dorb.com "Darrell" writes:
> Try something like passing a parm that allows access to
> an object in another module. The parameter approach also
> allows you to change the relationships of these modules
> at runtime.

I will probably do something like this -- I am thinking of re-writing
the software so that one module has a factory method in that creates 
instances of that module's classes; the factory method is then passed
as a variable to the other module, which calls it when it want to
instanciate classes in the 1st module.
 

-- 
Phil Hunt....philh at vision25.demon.co.uk




From bowman at montana.com  Sun Aug  1 10:42:24 1999
From: bowman at montana.com (bowman)
Date: Sun, 01 Aug 1999 08:42:24 -0600
Subject: How to find out the pid of a running program?
References: <7nucu2$11ko@enews1.newsguy.com> <37A31E4B.2CD53D40@ibm.net> <7o1989$pq7@enews2.newsguy.com>
Message-ID: <37A45CD0.93C2872E@montana.com>

Peter Torstenson wrote:
> 
> I want to find out the pid of another program than the one I give
> the command from!

How about os.popen("ps -ef") or whatever works on your OS. You can then
search
for 'pppd'  and get its pid. Not a great solution, as 'ps' flags and
formats vary across platforms. OTOH, is you needed to know the owner of
the process or other info, you'd get it all in one shot. 

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From fdrake at acm.org  Thu Aug 19 13:10:44 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 19 Aug 1999 17:10:44 GMT
Subject: Kudos to documentors: (popen2)
In-Reply-To: <37BC327A.3DA151EA@darmstadt.gmd.de>
References: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de>
	<37bc15c2@cs.colorado.edu>
	<37BC21D9.E19DBEA2@darmstadt.gmd.de>
	<14268.10767.95070.208607@weyr.cnri.reston.va.us>
	<37BC327A.3DA151EA@darmstadt.gmd.de>
Message-ID: <14268.14996.935343.952680@weyr.cnri.reston.va.us>

Robb Shecter writes:
 > That's what I was trying to say!  I think that my writing was not too
 > clear - I was trying to say that the Java and Python doc systems
 > should be used as examples for other languages.

  Thanks!

 > 3) I needed to know how to work with the return values.  The page says
 > "Returns (child_stdout, child_stdin)".  I got stuck here.  Without the
 > type, I didnt know how to read/write to them.  What I eventually did

  I've modified the documentation source to mention that they are
"file objects," which should be good enough for most regular Python
users.  Hyperlinking to the description of file objects will have to
come later (not impossible, but may require markup extension, which
I'd like to avoid at the moment for a variety of reasons).

 > It would have been great if their return types were listed, along with
 > hyperlinks to the appropriate docs.  But that wouldn't come for free
 > in a dynamically typed language, I guess.

  This depends on what you mean by "type"; in Python, it's reasonable
to call an interface (set of method signatures) a type.  No, it would
not be free.  I'd be more concerned about overpowering the user with
hyperlinks; if every reference to file-link objects was a hyperlink to 
the description of file-like objects, there'd be a lot of colorful
text in those documents!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From doug at tippett.com  Thu Aug 26 15:31:49 1999
From: doug at tippett.com (Doug Epps)
Date: Thu, 26 Aug 1999 12:31:49 -0700
Subject: help with AST and c/c++
Message-ID: <37C59625.77A17032@tippett.com>

hi,

does anyone have or know of any examples of using the 'AST' layer of
python from c/c++

specifically i'd like to parse my python code and then be able to see
what statements/parse-nodes
use what variables.

any help would be greatly appreciated 

yours truly

doug

ps - please respond via email as well, our news server is sometimes
wonky

-- 
Doug Epps -- Tippett Studio
510 649 9711 (voice)
510 649 9399 (fax)
doug at tippett.com



From jae at ilk.de  Mon Aug 23 21:17:31 1999
From: jae at ilk.de (Juergen A. Erhard)
Date: Tue, 24 Aug 1999 03:17:31 +0200
Subject: Man pages would be cool.
In-Reply-To: <37BE9EAB.CB7B1380@prescod.net> (message from Paul Prescod on
	Sat, 21 Aug 1999 12:42:19 GMT)
References: <37BC6986.A66B044E@starmedia.net>
				<199908191634.KAA01333@jhereg.perl.com>
				<14268.21198.934117.241668@anthem.cnri.reston.va.us>
				<14268.23007.239507.166209@weyr.cnri.reston.va.us>
		<y0j7lmq44wd.fsf@vier.idi.ntnu.no>
		<37BDD1B9.99588CB@prescod.net> <37BE24E0.79EA064E@montana.com> <37BE9EAB.CB7B1380@prescod.net>
Message-ID: <24081999.1@sanctum.jae>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

(A little Debian advertising is in order, I guess ;-)

In Debian potato[1], IDLE is found in the WindowMaker[2] root menu
under Apps/Programming.  (Debian 2.1 has python 1.5.1, there was no
idle).

Just as it appears in most of the other window managers that Debian
supplies.[2]

Bye, J

[1] potato is the unstable release, slated to be frozen by Nov 1st
(and to be released sometime in the next millenium probably ;-)

[2] Some are to weird to have IDLE in their root menu... twm, for
example (but then twm doesn't *have* a menu with any apps in the
default config ;-)

- -- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
         My WebHome: http://members.tripod.com/~Juergen_Erhard
            GNUstep - Free OPENSTEP (http://www.gnustep.org)
  "Windows NT" is an acronym for "Windows? No thanks." -- Russ McManus
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v0.9.10 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAjfB8qcACgkQN0B+CS56qs3o9wCeMC2krD7+anKpbzXFp6DAUxeh
VAUAoI1vY0A12aNiQFHmhIBocZo7Ylz8
=ICcs
-----END PGP SIGNATURE-----



From I.Clarke at strs.co.uk  Mon Aug 23 05:37:43 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Mon, 23 Aug 1999 10:37:43 +0100
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <37BEC755.645A350D@fnmail.com>
Message-ID: <37C11667.1BB653F8@strs.co.uk>

[snip]
> I suggest to treat the "bigger" guys in the computer community
> with more respect...
[snip]

And how do you define the "bigger" guy?  Because they write a book? 
Plenty of stupid people have written books.  Because they know alot
about Perl?  I know alot about distributed networking and Java, I don't
expect people to forgive me when I am rude or say stupid things because
of that.

I agree that T.C is a smart guy, but he lost my respect the day he
bombarded me with emails and then kilfiled my email address (despite my
responses being surprisingly polite given the circumstances).  If you
think that is bad what about the guy he has set up an automatic spam to
because he doesn't like his posting style.  I must say it was pretty
pathetic the way some C.L.P.M people tried to defend his actions because
they thought he was a "bigger guy".

Ian.



From herzog at online.de  Thu Aug 19 16:16:21 1999
From: herzog at online.de (Bernhard Herzog)
Date: 19 Aug 1999 22:16:21 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <19990819174800.D5227@xs4all.nl> <37bc313b@cs.colorado.edu>
Message-ID: <m3btc334ru.fsf@greebo.nodomain.de>

Tom Christiansen <tchrist at mox.perl.com> writes:

>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.python, 
>     Thomas Wouters <thomas at xs4all.nl> writes:

> :> 	>>> for c in ("weird", "bits"):
> :> 	>>> for c in ("weird"):
> 
> :Wait, wait, you're doing two different things here. First off, you called
> :'for' with different types of arguments, the first time with a tuple of
> :strings, the second type with just a string. 
> 
> Oh, come now, you're telling me I'm expected to count the items
> in the parens to know whether a split will happen.  Whatever. 
> It's not obvious, so it's a real gotcha.  This is not consistent.
> 
> :But they are definately not the same :)
> 
> But they look the same, which is wicked.
> 
[snip]
> 
> :"Yes, the trailing comma counts" :)
> 
> Speaking of inconsistencies!

The thing to note about tuples is that in a way it's not the parens that
make a tuple, it's the comma:

>>> t = 1, 2
>>> t
(1, 2)

The inconsistency is that () is an empty tuple, not that (1,) is a
one-element tuple. The parens are often needed because commas are also
used in list and dictionary literals and function calls and definitions,
among other places.

This is also why you can't slice dictionaries (if I understood what you
meant by slicing dicts in perl correctly). d[1,2] is equivalent to
d[(1,2)], i.e. it refers to the item with (1,2) as the key:

>>> d = {}
>>> d[1,2] = 'xyzzy'
>>> d
{(1, 2): 'xyzzy'}


-- 
Bernhard Herzog	  | Sketch, a drawing program for Unix
herzog at online.de  | http://www.online.de/home/sketch/



From MHammond at skippinet.com.au  Thu Aug 12 09:56:10 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 12 Aug 1999 23:56:10 +1000
Subject: Proper application of the buffer interface
References: <14248.63286.73014.204886@weyr.cnri.reston.va.us> <006501bedeee$33ef5c20$1101a8c0@bobcat>
Message-ID: <7ovim8$5b2$1@m2.c2.telstra-mm.net.au>

About a week ago, I wrote in message
<006501bedeee$33ef5c20$1101a8c0 at bobcat>...

>Just to extend my guesswork somewhat, there is a new built-in "buffer()"
>function.  This returns a "buffer" object.  I speculate this should be used
>in preference to Python strings when you have binary data.  As this buffer
>object supports the buffer interfaces, they are basically as functional as
>strings for this purpose, but clearly indicate the data is not really a
>string!  This appears to be more a matter of style, and also paves the road
>to Unicode - eg, it makes sense to convert any Python string object to
>Unicode, but not necessarily a binary buffer.

Just to prove it was guess-work, Bill Tutt indirectly pointed me at my error
here.

I may have given the impression that the buffer object returns a completely
new object - similar to a string.  The reason I may have given that
impression is because it is exactly what I meant.  However, it is wrong (and
I did know it was wrong <sigh>)

The "buffer()" built-in returns a read-only buffer for an _existing_ object.
It would only be an effective replacement for a string as a "byte buffer"
when coupled with an array object.

>>> import array
>>> a=array.array("b", range(ord('A'), ord('G')))
>>> b=buffer(a)
>>> b
<read-only buffer for fdcca0, ptr fdcc90, size 6 at fdccd0>
>>> b[:-1]
'ABCDE'
>>>

Just thought it worth a clarification.

As a side note, this also makes it far more difficult for extensions to
replace the use of strings with a different object for "byte buffers" - it
appears PyArray_New does not exist, so extension module writers are forced
to implement their own object if they wish to use the buffer interface, but
not use a PyString.

Mark.






From silver at mail.webspan.net  Thu Aug 19 19:12:54 1999
From: silver at mail.webspan.net (Andy Gaynor)
Date: Thu, 19 Aug 1999 18:12:54 -0500
Subject: 'if' as a _function_, in Scheme and Python
References: <7pi0kl$ko8$1@nnrp1.deja.com>
Message-ID: <37BC8F76.A836B0ED@mail.webspan.net>

Consider:

    (define (boolean->integer x)
      (if x 1 0))

    (define (if-function condition consequent-thunk alternate-thunk)
      ((vector-ref `#(,alternate-thunk ,consequent-thunk)
                   (boolean->integer condition))))

That is, `if' can well be thought of as syntactic sugar for a particularly
easy-to-optimize `case' expression.

Regards, [Ag]   Andy Gaynor   silver at mail.webspan.net



From kaz at ashi.FootPrints.net  Wed Aug 25 11:21:53 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Wed, 25 Aug 1999 15:21:53 GMT
Subject: xanadu and python
References: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com> <14276.2083.275770.273576@dolphin.mojam.com>
Message-ID: <slrn7s82hf.g2c.kaz@ashi.FootPrints.net>

On Wed, 25 Aug 1999 10:15:00 -0500 (CDT), Skip Montanaro <skip at mojam.com> wrote:
>Well, I get:
>
>    500 Server Error
>
>    The hard transfer limit for this user has been reached
>
>Not sure what that has to do with either Python or Xanadu...

Nothing. It has to do with the Slashdot effect. The story about Xanadu
being open sourced hit Slashdot last night.



From jbauer at rubic.com  Fri Aug 20 16:51:35 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Fri, 20 Aug 1999 20:51:35 GMT
Subject: Python+Apache+Oracle+???Persistent database connection???
References: <3F8BA119B48AD11197470008C7A493CB98B51D@HQOBGEX1>
Message-ID: <37BDBFD7.B8551572@rubic.com>

ScherBi at BAM.com wrote:
> You can acheive persistant database connectivity by converting
> your cgi apps to pcgi.  pcgi (persistant cgi), was developed by
> Jeff Bauer, and is available as a component of Zope 
> (http://www.zope.org).  You can use pcgi with or without Zope.

A minor correction:

The Persistent CGI protocol was already developmented by 
Digital Creations (when I was still in my nappies).  I'm 
merely its current custodian.

Jeff Bauer
Rubicon Reseach
"There ought to be limits to freedom."
http://www.gwbush.com



From python-list at teleo.net  Sat Aug 28 18:28:47 1999
From: python-list at teleo.net (Patrick Phalen)
Date: Sat, 28 Aug 1999 15:28:47 -0700
Subject: Returning none
References: <slrn7sgesi.lbd.thantos@brimstone.mecha>
Message-ID: <99082815301626.02087@quadra.teleo.net>

On Sat, 28 Aug 1999, Alexander Williams wrote:

> There are no procedures in Python.  There are only function calls (and
> Zuul).

Zuul?

Must be Dutch ...



From m.faassen at vet.uu.nl  Sun Aug 29 13:19:01 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 29 Aug 1999 17:19:01 GMT
Subject: Returning none
References: <slrn7sgesi.lbd.thantos@brimstone.mecha> <99082815301626.02087@quadra.teleo.net>
Message-ID: <7qbq25$kj6$1@newshost.accu.uu.nl>

Patrick Phalen <python-list at teleo.net> wrote:
> On Sat, 28 Aug 1999, Alexander Williams wrote:

>> There are no procedures in Python.  There are only function calls (and
>> Zuul).

> Zuul?

> Must be Dutch ...

It ain't Dutch. :)

Probably-some-PSU-codeword-ly yours,

Martijn




From news at dorb.com  Thu Aug 26 00:20:56 1999
From: news at dorb.com (Darrell)
Date: Thu, 26 Aug 1999 00:20:56 -0400
Subject: MSDOS window
References: <7q2brn$u7h$1@nnrp1.deja.com>
Message-ID: <Dg3x3.1590$v6.102977@typhoon.nyroc.rr.com>

Open a dos window.
>From the prompt, assuming that python.exe is in your path, type python.

Unless you install the windows specific version.
http://www.python.org/windows/win32all/

Little things like this can trip you up. Still remember how long it took me
to figure out that 'C' is case sensitive.

--
--Darrell
<henry299 at my-deja.com> wrote in message news:7q2brn$u7h$1 at nnrp1.deja.com...
> Hi, i'm new at this. I just downloaded version
> 1.5.2 for windows 95. When i double click on a
> program in the windows explorer, (example.py
> which is  print "This is some example text") an
> MS-DOS window opens and the text is displayed,
> but the window closes by itself in less than a
> second. Am i doing something wrong, and is there
> a way to keep the popup window open? Any help
> would be greatly appreciated.
>
> Thanks in advance,
>
> Matt
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.





From bowman at montana.com  Sun Aug  1 10:30:45 1999
From: bowman at montana.com (bowman)
Date: Sun, 01 Aug 1999 08:30:45 -0600
Subject: 1.5.2 for: else:
References: <Pine.LNX.4.04.9907271632450.9849-100000@emerald.netskate.ru> <199907272208.AAA06264@axil.hvision.nl> <7o0ipm$s18$1@ssauraab-i-1.production.compuserve.com>
Message-ID: <37A45A15.6B9323C5@montana.com>

T. C. Mits wrote:
> 
> ALL:
>     I'm a newbie so I may be out of line here, but the for: else: construct
> is pretty bad.  It just does not read well, and implies the incorrect
> operation; 

	I would agree that this use of 'else' is counterintuitive, though is
isn't quite as bad as Perl's use of 'continue'. Java's keyword,
'finally', would be another option, but that carries the meaning of a
block that is always executed.

	In practice, it doesn't seem to be necessary, so it can be ignored by
old C coders who can't quite remember how it works.

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From tchrist at mox.perl.com  Thu Aug 19 12:40:56 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 09:40:56 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <37BC1ED4.A29883BC@appliedbiometrics.com>
Message-ID: <37bc2588@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

>I had a look at this again. Despite the fact that the text in
>the above example is a bit misleading by "int of" where
>a round() takes place, which version of Python/architecture
>did you use? I get the expected result, see below.

Both SunOS and Linux round every other 5s up and down.  Somebody important
(IEEE I think) says you have to do this in order to regularize round-off
error.  Perl, Python, and C all behave identically on this on both those
platforms.  (I tested Perl and C on BSD, too, and they're the same.)
It's only Python's round() function that doesn't follow this.

1. C code 

    main() {
	printf("Float %g rounds to %.0f\n", 1.5, 1.5);
	printf("Float %g rounds to %.0f\n", 2.5, 2.5);
	printf("Float %g rounds to %.0f\n", 3.5, 3.5);
	printf("Float %g rounds to %.0f\n", 4.5, 4.5);
    }

    OUTPUT:

	Float 1.5 rounds to 2
	Float 2.5 rounds to 2
	Float 3.5 rounds to 4
	Float 4.5 rounds to 4

2. Perl code

    main() {
	printf("Float %g rounds to %.0f\n", 1.5, 1.5);
	printf("Float %g rounds to %.0f\n", 2.5, 2.5);
	printf("Float %g rounds to %.0f\n", 3.5, 3.5);
	printf("Float %g rounds to %.0f\n", 4.5, 4.5);
    }
    main();  # not called by default

    OUTPUT:

	Float 1.5 rounds to 2
	Float 2.5 rounds to 2
	Float 3.5 rounds to 4
	Float 4.5 rounds to 4

3. Python code: (did it really have to look so different? :-)

    # remember to un-indent this or it won't run
    print "Float %g rounds to %.0f" % (1.5, 1.5);
    print "Float %g rounds to %.0f" % (2.5, 2.5);
    print "Float %g rounds to %.0f" % (3.5, 3.5);
    print "Float %g rounds to %.0f" % (4.5, 4.5);

    OUTPUT:

	Float 1.5 rounds to 2
	Float 2.5 rounds to 2
	Float 3.5 rounds to 4
	Float 4.5 rounds to 4

You can see this in other situations, like working with pennies.
Imagine If you have extra half-pennies, like one and half cents,
two and half, there and half, etc:

    >>> print "You have $%0.2f bucks" % (2.5/100)
    You have $0.03 bucks
    >>> print "You have $%0.2f bucks" % (3.5/100)
    You have $0.04 bucks
    >>> print "You have $%0.2f bucks" % (4.5/100)
    You have $0.04 bucks

Strange but true.  It's supposed to even out the error.
Ask a numbers guy.  I'm just the messenger. :-)

As for the tuple thing, why can I say

    print "You have $%0.2f bucks" % (4.5/100)

How is that a singleton tuple?  Don't I have to use
a trailing comma?  It won't let me!  Why is 

    print "You have $%0.2f bucks" % (4.5/100,)

illegal, but 

    print "You have $%0.2f bucks" % ((4.5/100,) * 2)

mandatory?  Was this (elt,) thing just an, er, unforeseen design
misfeature? Yes, I tripped me up again. :-(

--tom
-- 
I have a different view of the world.  --Andrew Hume. Show&Tell '87



From fdrake at acm.org  Mon Aug 23 15:42:16 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 23 Aug 1999 19:42:16 GMT
Subject: More random python observations from a perl programmer
In-Reply-To: <Pine.LNX.4.10.9908201446540.22660-100000@atrus.jesus.cam.ac.uk>
References: <37bc1339@cs.colorado.edu>
	<m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk>
	<37bc4111@cs.colorado.edu>
	<m3n1vne72r.fsf@atrus.jesus.cam.ac.uk>
	<14269.22704.714782.33024@weyr.cnri.reston.va.us>
	<Pine.LNX.4.10.9908201446540.22660-100000@atrus.jesus.cam.ac.uk>
Message-ID: <14273.42008.836818.516697@weyr.cnri.reston.va.us>

Michael Hudson writes:
 > And so it is. I didn't think there'd so much hot air about the
 > non-existence of something that actually exists.

  Like Tom said, this is Usenet!  ;-)

 > I even learnt about the `-d' option from it. Can't see what it does

  Probably nothing if you don't set the right debugging flags in your
compile; it looks like legacy to me.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From Audun.Runde at SAS.com  Wed Aug  4 14:09:26 1999
From: Audun.Runde at SAS.com (Audun S. Runde)
Date: Wed, 4 Aug 1999 18:09:26 GMT
Subject: stuck on COM interface to Word
References: <37A60515.370B5085@sr.hp.com>
Message-ID: <FFyD3p.FEE@unx.sas.com>

Jim,

Jim Kerr <jbkerr at sr.hp.com> wrote in message
news:37A60515.370B5085 at sr.hp.com...
>   I'm trying to write a python script to extract some properties from a
set
> of Word documents. The attributes I'm interested in are things like
author's
> name, last print date, keywords, etc.

This should give you an idea, I hope:

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> from win32com.client import constants, Dispatch
>>> import string, os
>>> wd = Dispatch('Word.Application.8')
>>> wd.Documents.Open(r"c:\temp\test.doc")
<win32com.gen_py.Microsoft Word 8.0 Object Library.Document>
>>> doc = wd.ActiveDocument
>>> myprops = ("Author", "Subject", "Number of Words")
>>> for prop in myprops:
...  print prop, doc.BuiltInDocumentProperties(prop)
...
Author Audun S. Runde
Subject Test Subject
Number of Words 5
>>> print doc.CustomDocumentProperties("TestCustom")
Python Rules!
>>>


>   I tried using combrowse.py to figure out what the properties of the doc
are,
> but I couldn't find anything that looked relevant.
>
>   Is there any web-based info on COM interfaces to various programs? Or do
you
> have to buy Visual Basic books and translate for yourself?

To find out what to do here, you will have to cross over from Python to
Word/VBA. As another post mentioned, there is info about this in the online
Word help -- just make sure you explicitly installed the VBA help (it does
not come as part of a 'typical' install).

In your case, seach for BuiltInDocumentProperties in the online Visual Basic
Reference. To find the property names (i.e. 'Author', 'Subject', ...), seach
for DocumentProperty Object. These names also map to wdXXXX constants.

Generally, a good place to start is to turn on Word's Macro Recorder
(Tools->Macro-> Record New Macro...), do in Word what you'd like to
automate, then inspect the recorded macro for hints. In this case, however,
it doesn't work because accessing the Document Properties doesn't appear to
"do" anything in a macro.

A good place to look for more would be Microsoft Developer Network online at

    http://msdn.microsoft.com/default.asp

Hope this is useful :-)

Audun S. Runde
SAS Institute Inc.
Audun.Runde at SAS.com






From jlj at cfdrc.com  Tue Aug 24 09:59:26 1999
From: jlj at cfdrc.com (Lyle Johnson)
Date: Tue, 24 Aug 1999 13:59:26 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home>
Message-ID: <37c2a462.349531009@192.77.87.74>

>
>This is tcl/tk. Almost no difference. A bit more verbose, but there are
>numerous ways to shorten code, which I haven't used intentionally until
>I got tired and used a loop to produce four radiobuttons.
>

I agree that you're not saving yourself a lot of typing if you're only
going to do a Tk implementation of your GUI, or only a GTK
implementation, etc. But the point of this guy's "Parrot" is that you
can plug in different back-ends to generate, e.g. Tcl scripts or HTML
forms or C/GTK code or whatever.

I *personally* don't have a need to generate different implementations
of my GUIs in this way; but if I did I could see this tool being
useful.



From bruce_dodson at bigfoot.com  Mon Aug 30 21:59:43 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Mon, 30 Aug 1999 22:59:43 -0300
Subject: 'if foo' calls getattr?!?
References: <joe-300819991050234291@chinacat.salk.edu> <6ocvdDAPqsy3EwxA@jessikat.demon.co.uk>
Message-ID: <wHGy3.80681$jl.51638529@newscontent-01.sprint.ca>

If what you mean is really "is something" then "if not foo is None:" is what
you should say here.  However, I think getattr is only called if the method
isn't found in the dictionaries.  If that's true, I suggest defining
__nonzero__ to avoid those expensive getattr calls, so that when someone
does say "if foo:" performance will stay reasonable.  (Anyway, sometimes "if
foo:" is the right thing to say, since there are other ways for an object to
be untrue without being None.)

Robin Becker wrote in message >It seems that it tries to call
foo.__nonzero__()  and then foo.__len__()
>to test whether foo is true in some sense.
>
>I think these are standard behaviours for class instances. I think you
>should consider using if not foo is None: pass
>--
>Robin Becker




From vijaym at netscape.com  Fri Aug 20 02:08:29 1999
From: vijaym at netscape.com (vijay Mallela)
Date: Thu, 19 Aug 1999 23:08:29 -0700
Subject: pythonwin
Message-ID: <37BCF0DD.EB99F9F0@netscape.com>

does pythonwin has 1.5.2 v of python.. I am new to python and the
download page says that includes build 125.. can some one clarify
thx
v




From Jim.Crumpler at unicity.com.au  Fri Aug 13 08:22:22 1999
From: Jim.Crumpler at unicity.com.au (Jim Crumpler)
Date: Fri, 13 Aug 1999 13:22:22 +0100
Subject: L postfix char on longs - what is the best way to discard it.?
Message-ID: <934546409.21489.0.nnrp-14.d4e5b0e4@news.demon.co.uk>

Hi folks - I'm batting to find a nice clean way to remove the 'L' from long
repr strings .. (eg, my SQL queries end up like "SELECT * FROM blah WHERE
guff = 23423L"..)

I can fiddle with every bit of my code that attempts to display a long, such
as

query = "... WHERE guff = %s" % repr(long(x))[0:-1]

 -- its ugly when you use alot of longs, but does the job..

I can bugger around with long_format() to remove the postix 'L' altogether,
or build a % substitution to print a long without the L  - uglier for
distribution, but nice for writing code - I have no idea what repercussions
this has on the library code - probably bad.. I'm assuiming the actual
reason for the 'L' is so a string could be then fed back as a literal
without overflow the default integer conversion.  Couldn't the integer
conversion build a long on overflow, or would that get uglier?

Does anyone have some suggestions on better methods?

Thanks.

Jim Crumpler.






From x_mst at propack-data.com  Mon Aug 23 12:51:01 1999
From: x_mst at propack-data.com (Michael =?iso-8859-1?Q?Str=F6der?=)
Date: Mon, 23 Aug 1999 18:51:01 +0200
Subject: Sending a mail
References: <37C1704B.D1B173E5@daimi.au.dk>
Message-ID: <37C17BF5.6A055024@propack-data.com>

Peter Posselt Vestergaard wrote:
> 
> I want my script to send a mail with a specific subject. I've tried to
> import and use the smtplib, but I get the message that it doesn't exist

You should really try smtplib since it's portable.

> - maybe it isn't implemented on the server I'm using, and anyway - I
> can't specify a subject using the sendmail()-command...

Read the smptlib docs more carefully. You have to write the whole header
containing From:, Subject: etc. yourself.

> text='echo "hello"|mailto posselt at daimi.au.dk -s "important message"'
> os.system(text)
> 
> This works quite fine in my python interpreter, but not then I put it on
> the web.

This depends on your web server's OS and mail installation. Anyway you
have to ask your webmaster how to send e-mail from the shell.

> I tried to print out the standardoutput from os.system(text)
> and it was 256.

Did you really printed stdout? 256 seems to be the return code of the
os.system() call.

Ask your webmaster to install the missing smptlib.

Ciao, Michael.



From mcfletch at vrtelecom.com  Fri Aug 13 17:51:39 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Fri, 13 Aug 1999 21:51:39 GMT
Subject: Question about order in lists
In-Reply-To: <DH%s3.206$O02.12373@typhoon.nyroc.rr.com>
References: <DH%s3.206$O02.12373@typhoon.nyroc.rr.com>
Message-ID: <NDBBIAPHEKGOBIHCBLIIMEECCAAA.mcfletch@vrtelecom.com>

Or even:

>>> import operator
>>> l1=[5,3,7]
>>> l2=[8,6,3]
>>> map( operator.add, l1,l2)
[13, 9, 10]
>>>

Enjoy,
Mike

-----Original Message-----
From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
Behalf Of Darrell
Sent: August 13, 1999 4:59 PM
To: python-list at cwi.nl
Subject: Re: Question about order in lists


Worked for me.

>>> l1=[5,3,7]
>>> l2=[8,6,3]
>>> x=[]
>>> for i in range(len(l1)):
...     x.append(l1[i]+l2[i])
...
>>> x
[13, 9, 10]

--
--Darrell

Thorsten Z?ller <tzoeller at t-online.de> wrote in message
news:7p1v28$vcf$1 at news04.btx.dtag.de...
> Suppose you write a code as follows:
>
> x = []                                                # empty list
> list1 = [5, 3, 7]                                  # list containing any
> numbers
> list2 = [8, 6, 3]                                  # another list
containing
> any numbers
> for i in range(len(list1)):
>     x.append(list1[i-1] + list2[i-1])       # [5+8, 3+6, 7+3]
> ...
> x                                                        # print resulting
> list
>
> I'm trying to achieve that, having two lists of the same lenght containig
> any numbers, each component of the two lists is added to his pendant of
the
> other lists (i.e. the first component of list1 is added to the first
> component of list2 etc). The result is saved in another list x. But if I
run
> this code, I do not get [13, 9, 19] for x, but [10, 13, 9], i.e. the
> components are in a wrong order now. Why does the order change and how can
I
> avoid it?
>
> Thank you
> Thorsten
>
>






From greg.ewing at compaq.com  Thu Aug 19 00:29:52 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Thu, 19 Aug 1999 16:29:52 +1200
Subject: PIGgies?
References: <19990818171523.A22207@digex.net>
Message-ID: <37BB8840.E25D4EAE@compaq.com>

John Landahl wrote:
> 
> I later noticed that the term "PIGgies" is used elsewhere for
> Python user groups.  Am I the only Python enthusiast that has a problem with
> the term "PIGgies"?

It's the first time I've seen this term, and I've been
following this group for quite a while.

Maybe it would help if you spelled it Pyggies :-?

Greg



From thomas at xs4all.nl  Sun Aug 22 15:13:19 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Sun, 22 Aug 1999 21:13:19 +0200
Subject: Class Methods (as opposed to Instance Methods)
In-Reply-To: <7pjfhm$48p$1@feed.teaser.fr>
References: <7pjfhm$48p$1@feed.teaser.fr>
Message-ID: <19990822211319.B706@xs4all.nl>

On Fri, Aug 20, 1999 at 01:45:53PM +0200, Olivier Deckmyn wrote:

> Instance methods are easy to use:

> instance=MyClass()
> instance.myInstanceMethod()

> But here is how I would like to use class Methods :
> print MyClass.myClassMethod(2)
> (Note that there is no MyClass().xxx but well MyClass.xxx)
> I don't find any way to do that :(
> I know it exists class attributes, but it does not fit my needs :(

Cant be done(*), class functions expect a class instance as first argument at
all times. You can fake it by giving an unused instance (print
MyClass.myClassMethod(MyClass(),2) or by not using a class at all, but a
module (thus creating a seperate namespace)

file MyClass.py:
def myClassMethod(x):
	# etc
^D
file myprogram.py:

import MyClass
print MyClass.myClassMethod(2)

But in general it's best just to avoid using class methods and use either
general methods, module methods (if you aren't writing a module, it's worth
considering to make it a module after all ;-) or just instance methods.
Instance methods dont really have to _use_ the 1st argument (usually 'self')
they're called with.

(*) As far as I know. I've been proven wrong before.

Humbly-y'rs, Thomas.

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From bwarsaw at cnri.reston.va.us  Thu Aug 19 13:57:59 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 13:57:59 -0400 (EDT)
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
	<19990819174800.D5227@xs4all.nl>
Message-ID: <14268.17831.640587.502687@anthem.cnri.reston.va.us>

>>>>> "TW" == Thomas Wouters <thomas at xs4all.nl> writes:

    TW> There is some namemangling though, for datamembers/functions
    TW> that start with a single _, for hiding/overriding
    TW> purposes. (those mmebers get the classname encoded in, so it's
    TW> dependant on the class it was called on.)

The name manging rule in Python is actually "starts with double
underscore with no trailing underscores".  E.g. __mangled()

I like to call these `private' attributes, but they really aren't in
the C++/Java sense of privateness.  Because their names get mangled in 
a predicatable way, it's quite easy to get at them directly:

>>> class Shy:
... 	def __handsoff(self): print "please don't touch"
... 
>>> me = Shy()
>>> me.__handsoff()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
AttributeError: __handsoff
>>> me._Shy__handsoff()
please don't touch

If you want true privateness, you'll probably want to look at the
Bastion module, ExtensionClass, or some such.

Name mangling still serves an important purpose though.  It allows a
class to be designed for inheritance, with only the public and
`protected' interfaces needing documentation.  If I inherit from class
Shy above, my DerivedShy doesn't need to know that it's base class has
a private method called __handsoff() because my similarly named method
can't collide with it.

>>> class DerivedShy(Shy):
... 	def __handsoff(self): print "touch all you want"
..
>>> you = DerivedShy()
>>> you.__handsoff()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
AttributeError: __handsoff
>>> you._DerivedShy__handsoff()
touch all you want
>>> you._Shy__handsoff()
please don't touch

My personal coding convention has long been to use "starts with a
single underscore with no trailing underscores" to mean a `protected'
attribute; i.e. one that is intended to be used or overridden by
derived classes.  But there is no language support for this.

-Barry



From fredrik at pythonware.com  Thu Aug 26 08:47:25 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 26 Aug 1999 12:47:25 GMT
Subject: Documentation string
References: <7q3c88$k8m$1@nnrp1.deja.com>
Message-ID: <005f01beefc1$26c0f8e0$f29b12c2@secret.pythonware.com>

sp00fD <sp00fD at yahoo.com> wrote:
> Can someone tell me why the
> 
> def foo(self):
> """A method"""
> 
> won't work?

a docstring is a actually a statement (to be
exact, it's an expression statement with no
side effects).

in other words, docstrings have to be properly
indented, just like all other statements in the
function body.

</F>




From tzoeller at t-online.de  Fri Aug 13 16:29:50 1999
From: tzoeller at t-online.de (Thorsten Zöller)
Date: Fri, 13 Aug 1999 22:29:50 +0200
Subject: Question about order in lists
Message-ID: <7p1v28$vcf$1@news04.btx.dtag.de>

Suppose you write a code as follows:

x = []                                                # empty list
list1 = [5, 3, 7]                                  # list containing any
numbers
list2 = [8, 6, 3]                                  # another list containing
any numbers
for i in range(len(list1)):
    x.append(list1[i-1] + list2[i-1])       # [5+8, 3+6, 7+3]
...
x                                                        # print resulting
list

I'm trying to achieve that, having two lists of the same lenght containig
any numbers, each component of the two lists is added to his pendant of the
other lists (i.e. the first component of list1 is added to the first
component of list2 etc). The result is saved in another list x. But if I run
this code, I do not get [13, 9, 19] for x, but [10, 13, 9], i.e. the
components are in a wrong order now. Why does the order change and how can I
avoid it?

Thank you
Thorsten





From m.faassen at vet.uu.nl  Wed Aug 11 04:00:31 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 11 Aug 1999 08:00:31 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B02965.B1FE0E8D@prescod.net> <934326002snz@vision25.demon.co.uk>
Message-ID: <7oraiv$nuo$1@newshost.accu.uu.nl>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
[snip]
> I am intending to do something like this. For example:

>    sub.font = "Helvetica"

> would mean: all components under the current component default to
> using the Helvetica font.

You may want to look at Digital Creation's acquisition for this. I saw
it mentioned once that this might be useful for GUI's too. It's used
all over Zope and in Zope it's very powerful.

A page on acquisition:

http://www.digicool.com/releases/ExtensionClass/Acquisition.html

> Similarly:

Actually this is quite different from the previous example. You wouldn't
use acquisition for this one, I think. :)

>    all.button.bcol = "111133"

> would make the background colour of all buttons default to dark blue.

Regards,

Martijn




From sjoshi at ingr.com  Thu Aug 12 12:40:01 1999
From: sjoshi at ingr.com (sjoshi at ingr.com)
Date: Thu, 12 Aug 1999 16:40:01 GMT
Subject: Reading/Extracting Zipped Files ??
Message-ID: <7outd0$c05$1@nnrp1.deja.com>

Is there a module that will let me read .zip files and extract
selected files from the archive.

thanks
Sunit


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From aahz at netcom.com  Thu Aug 19 14:59:11 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Aug 1999 18:59:11 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <37bc2ee5@cs.colorado.edu> <7phfe0$19so$1@nntp6.u.washington.edu> <37bc487e@cs.colorado.edu>
Message-ID: <7phk5v$ot1@dfw-ixnews8.ix.netcom.com>

In article <37bc487e at cs.colorado.edu>,
Tom Christiansen  <tchrist at mox.perl.com> wrote:
>
>The Prisoners of Bill gets HTML, regular Unix users get manpages,
>and anyone can have simple text.  And everyone is happy.

Excuse me, but I do much of my Python development on Linux and Solaris,
where Lynx (with its ease of navigation) does a much better job on
documentation than man pages -- for me.  (I mentioned this in another
post, but I figured sticking in the OS references was worth a repeat.)
I can't tell you how many times I've gotten pissed at having one Perl
man page reference another with no direct mechanism for getting there.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From carlos.ors at usa.net  Thu Aug 12 03:46:45 1999
From: carlos.ors at usa.net (Carlos Kyoku)
Date: Thu, 12 Aug 1999 07:46:45 GMT
Subject: I need help
Message-ID: <19990812074645.11685.qmail@www0s.netaddress.usa.net>

Hi!

I'm trying to access COM from jpython but I don't know how. I tried to access
using win32com modules which belongs python but I can't neither.

How can I acces COM from jpython?

                                         A lot of thanks.



From amk at mira.erols.com  Mon Aug  9 21:18:34 1999
From: amk at mira.erols.com (A.M. Kuchling)
Date: 10 Aug 1999 01:18:34 GMT
Subject: gzip bug?
References: <7onios$4dt$1@nnrp1.deja.com>
Message-ID: <slrn7quvu0.1eo.amk@207-172-38-11.s11.tnt8.ann.va.dialup.rcn.com>

On Mon, 09 Aug 1999 21:55:46 GMT, 
	catlee at my-deja.com <catlee at my-deja.com> wrote:
>gzip file objects don't seem to close themselves automatically.  When I
>do this:

Because they don't have a __del__ method, so when you're writing the
output file isn't closed (and the checksum isn't written).  Try this
patch:

Index: gzip.py
===================================================================
RCS file: /projects/cvsroot/python/dist/src/Lib/gzip.py,v
retrieving revision 1.12
diff -C3 -r1.12 gzip.py
*** gzip.py	1999/04/12 14:34:16	1.12
--- gzip.py	1999/08/10 01:16:51
***************
*** 251,256 ****
--- 251,259 ----
              self.myfileobj.close()
              self.myfileobj = None
  
+     def __del__(self):
+         if self.myfileobj is not None: self.close()
+         
      def flush(self):
          self.fileobj.flush()
  

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
The quality of what is said inevitably influences the way in which it is said,
however inexperienced the writer.
    -- Robertson Davies, _A Voice from the Attic_



From news at 1492.net  Thu Aug  5 09:33:19 1999
From: news at 1492.net (news at 1492.net)
Date: Thu, 05 Aug 1999 13:33:19 GMT
Subject: Movie Website!!
Message-ID: <05089908.4044@1492.net>

New Entertaining Movie Game Website!

http://www.moviedoubletakes.com

Totally free!! Test your knowledge!!




,4LF(



From tim_one at email.msn.com  Sun Aug 15 22:31:18 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 15 Aug 1999 22:31:18 -0400
Subject: C's isprint() concept?
In-Reply-To: <7p7cuc$plr@dfw-ixnews5.ix.netcom.com>
Message-ID: <000201bee78f$6bbf6ec0$f22d2399@tim>

[Jeff Blaine, wants to
> ... replace all non-printable characters in a string
> with a single space

[Aahz Maruch]
> Oh, sure.  What isn't clear from the way you write this is whether
> a run of multiple non-printable characters should be replace with
> a single space.  Here's a regex that I created recently as part of
> some code to detect binary documents:
>
>    # Everything that isn't CR/LF or 7-bit normal characters
>    reBinary = re.compile ( r'[^\r\n\x20-\x7F]' )

Take a look at string.translate.

> ...
> Don't remember offhand what isgraph() does, but it occurs to me that
> it would be useful to have a character class string.printable.
> Tim?  Guido?

I can't speak for Tim, but Guido would say "be my guest -- define whatever
character classes you want".  I think he doesn't want to pee away time
arguing over e.g. whether \x7F is "normal" or not <wink>.

unicode-is-gonna-be-sooooo-much-fun-ly y'rs  - tim





From godzilla at netmeg.net  Sun Aug 22 12:47:13 1999
From: godzilla at netmeg.net (Les Schaffer)
Date: Sun, 22 Aug 1999 16:47:13 GMT
Subject: More random python observations from a perl programmer
In-Reply-To: <00c601beecbc$15ac4210$f29b12c2@secret.pythonware.com>
References: <000001beec6d$9753e240$542d2399@tim>
	<m2iu677ujv.fsf@netmeg.net>
	<00c601beecbc$15ac4210$f29b12c2@secret.pythonware.com>
Message-ID: <14272.10641.952348.992882@gargle.gargle.HOWL>

/F said:

> certain well-known Perlers sometimes like to pretend that the
> mathematical meaning is the only valid one, but that's not entirely
> true...

okay. but what is the significance of __this__ orthogonal in the
statement:

"The builtins strive to stick to those type/operator combos that make
plain sense; Guido has never been a fan of gratuitous orthogonality."

does it mean simply that guido et al believe not all type/operator
pairs need coverage in the builtins? something like "we dont feel the
need to span all of type/operator space since the phase space of
python coder trajectories seems well bounded by the span of existing
pairs"

les

-- 
____        Les Schaffer              ___| --->> Engineering R&D <<---
Theoretical & Applied Mechanics          |  Designspring, Inc. 
Center for Radiophysics & Space Research |  http://www.designspring.com/
Cornell Univ.  schaffer at tam.cornell.edu  |  les at designspring.com



From tismer at appliedbiometrics.com  Fri Aug 13 05:01:24 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Fri, 13 Aug 1999 09:01:24 GMT
Subject: Macro preprocessor.
References: <slrn7r291c.137.kaz@ashi.FootPrints.net>
		<37B2B71C.E380A71@appliedbiometrics.com> <y0joggdf71f.fsf@drue.idi.ntnu.no>
Message-ID: <37B3DEE4.55E12565@appliedbiometrics.com>


"Magnus L. Hetland" wrote:
> 
> Christian Tismer <tismer at appliedbiometrics.com> writes:
> 
> > Kaz Kylheku wrote:
> > [description of a cute macro processor skipped]
> >
> > > Okay, enough. :) I'll probably get hate mail now about being off topic.
> >
> > No hate mail, but some thoughts:
> [...]
> 
> > There are a few cases where I admit that I generate scriptlets
> > on-the-fly using the formatting operator.
> 
> Now you've piqued my interest... How do you do this, and in which
> cases (for what purposes)? Do you create strings that you execute, or
> what? (Sounds like a cool thing to do ;)

I wrote a helper class which fakes dynamic classes which
can be pickled and stored in database fields.
This was used for a glass fiber network simulation project, where
networks of nodes with lots of properties and direct or symbolic
references to other nodes have to be stored, edited and updated.

The first version of this used exec and formatting to generate
classes dynamically and provide attributes and methods.

Later I could replace almost everything with calls to the "new"
module, and dynnamic assignment of functions as methods.

This is a case where classes and instances are not so sharply
distinguished, it falls into the class of problems where
metaclasses are handy. One simulation defines a special
configuration, gathers data and is saved as an object in
the database, this makes it like an instance. But other
simulations can refer to it, use it as a building block
and inherit from it, which makes it look like a class.

This is a proprietary project, so I can show just some
excerpts from an outdated version of the source files.
System design has been completely redefined, so this is
history.

Here is a dynamic module generator:

----------------------------------------------------------------
#######################################################

# provide a virtual module which has no true file

DYN_MODULE_NAME = "dynamic node classes"

def __virtual_setup():
    import new, sys
    vm = new.module(DYN_MODULE_NAME)
    sys.modules[DYN_MODULE_NAME] = vm
    
__virtual_setup()
del __virtual_setup

DYN_MODULE = __import__(DYN_MODULE_NAME)

# an alternative could be to replace the standard __import__
# function with one that can behave as expected.
# Anyway, this works.

class NodeGenerator: # Generates *no* code to declare a new class
on-the-fly
    def __init__(self, base, tag):
        import new
        self.nodeLabel = cname = base + "_" + tag # name of the class to
generate
        self.classDef = cl = new.classobj(cname, (globals()[base],), {})
        def __init__(self, id):
            self.__class__.__bases__[0].__init__(self,id)
            # which has a slight semantical difference. Guess about it!
        cl.__init__ = __init__ # function becomes an unbound method
        cl.name = tag

        # Initialize empty parameter and port defs
        cl.parameterDefs = {}
        cl.portDefs = {}

        # provide a virtual module name
        cl.__module__ = DYN_MODULE_NAME
        
    def address (self, addr):
        # Record the persistent address that the class has
        self.classDef.address = addr

        
    def defParameter (self, name, type, default, unit, range, desc):
        # Record a parameter definition
        param = {}
        param[ "type" ] = type
        param[ "default" ] = default
        param[ "unit" ] = unit
        param[ "range" ] = range
        param[ "desc" ] = desc
        
        self.classDef.parameterDefs[ name ] = param
        

    def defPort (self, name, type, direction, cardinality, desc):
        # Record a port definition
        port = {}
        port[ "type" ] = type
        port[ "direction" ] = direction
        port[ "cardinality" ] = cardinality
        port[ "desc" ] = desc
        
        self.classDef.portDefs[ name ] = port


    def declareNode(self):
        # just return the ready thing
        return self.classDef

#######################################################
----------------------------------------------------------------

and here a factory for abstract nodes:

----------------------------------------------------------------

class AbstractNodeFactory:
    def __init__(self):
        self.knownNodes = {} # The known types of AbstractNodes keyed by
addresses
        
    def getNode(self, address):
        # Is the desired node known?
        if address not in self.knownNodes.keys():
            # No => generate on-the-fly
            generator = NodeGenerator( "AbstractNode", "SomeShortLabel_"
+ address )

            # Some dummy stuff
            generator.address( address )

            generator.defParameter( "Weight", "float", "42.0", "lbs",
"41.9-42.1", "The weight of the brain" )

            generator.defPort( "Eye", "optical", "input", "2", "Optical
input" )
            generator.defPort( "Ear", "acustical", "input", "2",
"Acustical input" )
            generator.defPort( "Mouth", "acustical", "output", "1",
"Acustical output" )

            # Declare the class and get to know it            
            generator.declareNode()            
            self.knownNodes[ address ] = generator.classDef # was
generatedClass
            # store in the dynamic module as well
            node = self.knownNodes[ address ]
            setattr(DYN_MODULE, node.__name__, node)
            # keep a reference to the factory instance as well
            node.Factory = self

        # Return the class object
        return self.knownNodes[ address ]
        
    # pickling support.
    # we don't store the generated classes, but the parameters
    # needed to restore them.
    # therefore, the __getstate__/__setstate__ method
    # avoid to make use of the __dict__
    def __getstate__(self):
        return self.knownNodes.keys()
        
    def __setstate__(self, state):
        self.__init__()
        map(self.getNode, state)

class persistency:
    """wrap a node instance suitable for pickling"""
    def __init__(self, node_inst):
        factories = {}
        instances = {} # maybe we must use id() in the future
        todo = [node_inst]
        while todo:
            lookat = todo[0]
            del todo[0]
            for port in lookat.ports.values():
                for conn in port.values():
                    inst = conn[0]
                    if not instances.has_key(inst):
                        todo.append(inst)
                        instances[inst] = 1
                        factories[inst.Factory] = 1
        self.factories = factories.keys()
        self.instances = instances.keys()
        self.saveset = (self.factories, node_inst)
        
    def __getstate__(self):
        return self.saveset
        
    def __setstate__(self, state):
        factories, instance = state
        self.__init__(instance)

#######################################################


-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From jbauer at rubic.com  Sat Aug 21 13:35:28 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Sat, 21 Aug 1999 17:35:28 GMT
Subject: ANNOUNCE normalDate 0.7
Message-ID: <37BEE360.2D96EB9D@rubic.com>

Hi folks.

I don't keep a list of people who use NormalDate, a simple
Python date class, but Dan Winkler reminded me today that
I neglected to put in __add__, __sub__ methods.  For anyone
using it, the latest version is available:

  http://starship.python.net/crew/jbauer/normalDate/

>From the docstring:

    NormalDate is a specialized class to handle dates without
    all the excess baggage (time zones, daylight savings, leap
    seconds, etc.) of other date structures.  The minimalist
    strategy greatly simplifies its implementation and use.

    Internally, NormalDate is stored as an integer with values
    in a discontinuous range of -99990101 to 99991231.  The
    integer value is used principally for storage and to simplify
    the user interface.  Internal calculations are performed by
    a scalar based on Jan 1, 1900.

For people who are serious about their date calculations, I'd
recommend Marc-Andre Lemburg's mxDateTime:

  http://starship.python.net/crew/lemburg/mxExtensions.html

For people who aren't so serious about their date calculations,
check out:

  http://www.pauahtun.org

The Python distribution also includes a date class, which would
have saved me time, had I bothered to check.

  Demo/classes/Dates.py

---
Jeff Bauer
Rubicon Research



From philh at vision25.demon.co.uk  Mon Aug 16 18:12:49 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 16 Aug 99 22:12:49 GMT
Subject: function which knows who has called it.
Message-ID: <934841569snz@vision25.demon.co.uk>

Is if possible in Python to have a function that prints the name
of the function that called it. (I'm thinking of using this in
some debugging code I want to write).

-- 
Phil Hunt....philh at vision25.demon.co.uk




From cbbrowne at news.hex.net  Mon Aug  9 20:56:48 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Tue, 10 Aug 1999 00:56:48 GMT
Subject: looking for py/database ideas
References: <m2so5slkdj.fsf@netmeg.net>
Message-ID: <kNKr3.879$gO1.33509@news2.giganews.com>

On Mon, 09 Aug 1999 21:22:07 GMT, Les Schaffer <godzilla at netmeg.net> wrote:
>a business/suit type friend of mine apprached me the other day about
>helping him port an existing commercial Paradox database app to
>something ""more modern"".
>
>having vanishingly small database experience, i proceeded first to
>browse the bookshelves for tips, and then after asking around some
>amongst software friends, have learned that using MS Access / DAO /
>VBA is considered de rigeur these days.
>
>needless to say, this has induced derigor mortis in me. 
>
>so i am fishing for ideas on other ways to build database
>applications, and having just built a 10k line engineering simulation
>app in python last year, i am interested in hearing if pythoners have
>ideas for implementing comerical-type databases in python.
>
>my buddy doesnt know what size market he has for this yet, but it will
>initially be aimed at local municipalities in the US (small to medium
>size towns and cities) who need to keep track of things (i shouldnt
>say what things yet) which live in that municipality. as such, we are
>not talking about high-availablity requirements -- wont be more than a
>handful to tens of people (this is high) accessing the database at any
>one time.
>
>web access to db MIGHT be important if people in the field needed to
>add data to the DB over a network connection (again, not a lot at one
>time). 
>
>the main concern over the existing Paradox DB is upgrade to what my
>friend thinks is a more modern db system -- thats his opinion, i have
>none since i know nothing about this -- whatever that means. i think
>he thinks moving to a newer db system would allow for better solutions 
>for db needs, and would move away from paradox which he thinks is a
>dying db system. 
>
>so..... i smell an opportunity to point him towards alternative
>solutions. 
>
>can things be done with python to create databases with a decent GUI
>interface and Web access (err, not MS access) -- i am playing with
>Zope a lil right now -- and SQL and all those goodie buzzwords. and
>how hard would it be to integrate a python-esque type solution with
>existing standard off-the-shelf systems running in other departments
>(in other words, if another db system in that municipality used, say,
>Acceess, could our solution be --- uhhh -- integrable with that system 
>(boy do i need some buzzwords). 
>
>just looking for some pointers to help my friend who, you may not
>believe this from reading the above, is far more clueless than me.

I've got some material at the URL below on a variety of approaches
that are being taken as alternatives to the "Embedding it In MS
Access" application development approach.

Rather than being Python-centric, it's somewhat Linux-centric.

If it has useful URLs, so be it; if there are constructive comments to
be made on it, email is fine; if there are interesting approaches and
relevant URLs, that would be welcome.

By the way, the term "integrable" is often used by mathematicians, but
almost never by "systems integrators."

-- 
Rules of the Evil Overlord #55. "I will maintain a healthy amount of
skepticism when I capture the beautiful rebel and she claims she is
attracted to my power and good looks and will gladly betray her
companions if I just let her in on my plans." 
cbbrowne at ntlug.org- <http://www.ntlug.org/~cbbrowne/framewrk.html>



From tchrist at jhereg.perl.com  Thu Aug 19 13:50:58 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Thu, 19 Aug 1999 11:50:58 -0600
Subject: More random python observations from a perl programmer 
In-Reply-To: Skip Montanaro's message of Thu, 19 Aug 1999 12:50:11 -0500 (CDT) .
             <14268.16208.414978.281563@dolphin.mojam.com> 
Message-ID: <199908191750.LAA03193@jhereg.perl.com>

>As for man pages, they aren't the only valid way to document reference
>material.  Most Windows and Mac people wouldn't even know what you were
>talking about.  From what I can tell by traffic on c.l.p, the bulk of the
>Python user base uses Windows these days anyway, so man pages are probably
>not a good first choice of documentation format, being largely unavailable
>to the unwashed masses.  HTML seems to be pretty ubiquitous these days,
>however.

Sigh.  And you wonder why folks say that alienating Unix programmers
has never been something Python has tried to avoid? :-(

--tom



From spencerm at cruzio.com  Wed Aug 18 12:36:05 1999
From: spencerm at cruzio.com (Michael Spencer)
Date: Wed, 18 Aug 1999 09:36:05 -0700
Subject: Help!  Dialling on modem
Message-ID: <7pemm3$fd5@enews3.newsguy.com>

Have looked thru docs & examples on networking. Lots of info on using
sockets and TCP. But nothing on how how to cause the modem to dial out.

Lots of good reasons for not using M$ TAPI.

Can anyone tell me where to find sample code on using modem to dial a
telephone number under Win32?

Thanks

Mike





From mcgrail at bard.edu  Fri Aug 13 16:22:36 1999
From: mcgrail at bard.edu (Robert W. McGrail)
Date: Fri, 13 Aug 1999 16:22:36 -0400
Subject: Lambda Functions (fwd)
References: <xpolakis-2708561956530001@athe530-a072.otenet.gr>
Message-ID: <37B47E8C.787801EF@bard.edu>


> Hi All--
> I'm working on a book about the Python programming language.  Python
> allows the use of lambda functions, small anonymous functions that can
> speed up processing because they are effectively inlined, and often
> require less typing than conventional function definitions.
>
> Why lambda?  I realize that the term has come from functional
> programming languages, and Python's inventor explicitly took lambda
> from Scheme, but I'm interested in the history of the term.  Why that
> particular letter?  I assume it has something to do with the lambda
> calculus, but my ignorance runs deep in that area too.

The lambda calculus is the reason as you suspect.  This makes sense since the
core of functional abstraction in Scheme (or more generally Lisp) is based on the
lambda construct in the (call-by-value) lambda calculus.  The question remains as
to the use of lambda in the lambda calculus.  If my sources are correct the use
of lambda results from the limitations of typesetters in the time of the lambda
calculus' creator, Alonzo Church.  It seems that Church wanted to use a bar above
a variable to denote that the particular variable was bound in the body of a
lambda expression.  However, at some point the bar was replaced by another symbol
(I forget which) preceding the binding instance of the variable.  This symbol
eventually was replaced by a lambda.

Of course I am missing some details here.

--
Best Regards,
  Bob McGrail

****************************************************

By virtue knowledge

****************************************************
Bard College
Division of Natural Science and Mathematics
P.O. Box 5000
Annandale-on-Hudson, NY 12504
(914)758-7265

mcgrail at bard.edu
http://inside.bard.edu/~mcgrail





From Dan at Grassi.com  Sun Aug 15 13:28:41 1999
From: Dan at Grassi.com (Dan)
Date: Sun, 15 Aug 1999 13:28:41 -0400
Subject: Mac ODBC
Message-ID: <199908151727.KAA24472@mb3.mailbank.com>

Marc-Andre Lemburg wrote:

>What database are you interfacing to?

Sybase & Oracle.

>If you can, you
>should link against the ODBC manager if available.

I will check into this.  I am using the Intersolv ODBC driver at the 
moment, I do not know if their interface is the same as the Apple 
(unsupported/obsolete Visigenics) interface.  Time will tell.

My time frame is during September.

Dan, Dan at Grassi.com



From a8gnt69 at aol.com  Mon Aug 30 11:53:26 1999
From: a8gnt69 at aol.com (a8gnt69 at aol.com)
Date: Mon, 30 Aug 1999 09:53:26 -0600
Subject: Geat XXX 50791
Message-ID: <300899095326@aol.com>

lolitas:

http://207.240.225.250


3a/-]=[C



From donn at u.washington.edu  Thu Aug  5 11:52:08 1999
From: donn at u.washington.edu (Donn Cave)
Date: 5 Aug 1999 15:52:08 GMT
Subject: Problems with sockets on Win NT
References: <7oaibf$763$1@nnrp1.deja.com>
Message-ID: <7ocbv8$t3q$1@nntp6.u.washington.edu>

mkrishnam at my-deja.com writes:
| I'm trying to exchange datagram (UDP) packets
| between a unix machine and a Win NT machine.
|
| The UDP packets sent from Win NT machine are
| received properly on unix machine. However UDP
| packets sent from unix machine is not at all
| received on WinNT machine.
|
| I'm using recvfrom method to receive the packets.
|
| The code I'm using is as below.

In this code, you seem to be using recv() instead of recvfrom().
Maybe you could follow up with some clarification.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu
-----------------------------
| # Code on Win NT machine
|
| from socket import *
|
| my_sock = socket(AF_INET, SOCK_DGRAM)
| my_sock.bind('localhost', 6123)
|
| while(1):
|         test = my_sock.recv(1024)
|         print "got it!"
|
| # Code on Unix machine
|
| from socket import *
|
| my_sock = socket(AF_INET, SOCK_DGRAM)
| my_sock.bind('localhost', 6440)
|
| dst_addr = 'leto',6123
|
| while (1):
|         print my_sock.sendto("Hello", dst_addr)



From philh at vision25.demon.co.uk  Mon Aug 16 17:42:44 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 16 Aug 99 21:42:44 GMT
Subject: [NOISE] [OFF-TOPIC] [BUT IMPORTANT] Re: Why use Python when we've got Perl?
References: <000301bee78f$6cd39de0$f22d2399@tim> <wmc4shzadhq.fsf@balard.polytechnique.fr> <m3n1vr3bw6.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <934839764snz@vision25.demon.co.uk>

In article <m3n1vr3bw6.fsf at atrus.jesus.cam.ac.uk>
           mwh21 at cam.ac.uk "Michael Hudson" writes:
> Jerome Kalifa <Jerome.Kalifa at polytechnique.fr> writes:
> [schnupp]
> > Indeed, the comparison is silly, but for a different reason : Asian
> > food, European food... this doesn't mean anything at all! What's in
> > common between Japanese, Chinese and Cambodian foods, or between
> > German, English, Spanish and Italian foods? Nothing. 
> 
> Well, they're all carbon based, genereally.
> 
> The funniest part of the perl/Python flame war is how similar
> languages are, especially when compared with stuff like lisp, ml or
> haskell.

Yep, Perl and python are semantically quite close (IMO). The difference
is Python has sensible syntax.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From earlybird at mop.no  Thu Aug  5 09:37:13 1999
From: earlybird at mop.no (Alexander Staubo)
Date: Thu, 5 Aug 1999 15:37:13 +0200
Subject: International characters and Python's string functions on Linux
Message-ID: <MPG.1213b1f2ede568ee989681@news.mop.no>

I'm running Python on Slackware Linux, and it does not seem to provide a 
full international character set in the string.uppercase and 
string.lowercase constants: only A-Z, which isn't enough.

Having studied the string module code a bit, I realize that Python is at 
the mercy of the C API and its string/locale functions, so I'm thinking I 
must probably configure Linux for a specific locale. Right? If so, how?

-- 
Alexander Staubo             http://www.mop.no/~alex/
"It has taken the planet Earth 4.5 billion years to discover it is
4.5 billion years old." --George Wald



From effbot at hotmail.com  Thu Aug  5 10:19:04 1999
From: effbot at hotmail.com (eff-bot script)
Date: Thu, 05 Aug 1999 14:19:04 GMT
Subject: Python GUI
Message-ID: <19990805141904.115.qmail@hotmail.com>

> > Is there a Python GUI builder out there anywhere?
>
>Not that I'm aware of.
>
>Anyone know of one in-the-works?

http://www.pythonware.com/products/works/


_______________________________________________________________
Get Free Email and Do More On The Web. Visit http://www.msn.com



From lwilson at address.com  Thu Aug 26 21:33:05 1999
From: lwilson at address.com (lwilson at address.com)
Date: 27 Aug 1999 01:33:05 GMT
Subject: Free Emachine eOne and Palm Pilot V
Message-ID: <7q4psh$jbj$4710@news.cadence.com>

Visit http://www.address.com and enter to win.   Tell them lwilson at address.com sent you!!!




From furcat at hawaiian.net  Sun Aug 29 01:07:19 1999
From: furcat at hawaiian.net (furcat at hawaiian.net)
Date: Sat, 28 Aug 1999 22:07:19 PDT
Subject: stress reduction video
Message-ID: <37c8b4ea@news.hawaii.rr.com>

MAHALO ENTERPRISES, HAWAII
DEDICATED TO BRINGING YOU AND YOUR FAMILY 
QUALITY PRODUCTS AT FANTASTIC DISCOUNTED PRICES.

LIVE LONGER,  
MAKE MORE MONEY, 
HAVE A BETTER RELATION WITH YOUR LOVE ONES, 
HAVE A BETTER SEX LIFE, 
MAINTAIN A POSITIVE MENTAL ATTITUDE 
CREATIVITY WILL  REACH NEW HIGHS
KEEP YOUR BLOOD PRESSURE WAY DOWN
RELAX YOUR NERVES
GET FOCUSED ON YOUR FUTURE
LEARN TO ELIMINATE THE NEGATIVES
AND STRENGTHEN THE POSITIVES
POTENTIAL OF SAVING THOUSANDS OF DOLLARS$$$ 
ON YOUR MEDICI AL BILLS

ALL THIS AND MUCH MORE IS YOUR NOW IS THE AWARD WINNING VIDEO
                                "DREAMS OF HAWAII"

MEDICIAL RESEARCH HAS NOW PROVEN THE IMPORTANCE OF A GOOD BALANCE IN YOUR DAILY LIFE WITH THE ELIMINATION OF STRESS. 

STRESS KILLS AND WEARS YOU DOWN FASTER THAN ANYTHING. 

MAKING TIME FOR YOURSELF IN A RELAXING AND COMFORTABLE ENVIRONMENT WILL ONLY ENHANCE YOUR LIFE.  

DREAMS OF HAWAII VIDEO BRINGS THE MOST UNIQUE AND BEAUTIFUL ENVIRONMENT IN THE WORLD INTO YOUR HOME FOR YOU AND YOUR FAMILY AND FRIENDS.
THIS IS A VIDEO THAT YOU PLAY OVER AND OVER AGAIN FOR YEARS AND YEARS.  

YOU WILL BE AMAZED THAT JUST A FEW MINI TUES WILL HELP YOU WITH ALL THE ABOVE MENTION CATEGORIES AND MORE.

YOU MAY ENJOY THE MUSIC ON THE TAPED OR PLUG IN YOUR FAVORITE MUSIC AND TURN OF THE SOUND ON THE VIDEO.

Check out our web site to view movie clips and full-size 
sample images from the video! 
http://www.video-hawaii.com/dreams.html

When you're ready to order, use this special order form 
to get your 25% savings and free limited-edition video: 
http://www.video-hawaii.com/special_order.html


HERE ARE SOME OF THE FEEDBACK WE ARE GETTING FROM THE PEOPLE THAT HAVE PURCHASED DREAMS OF HAWAII

LOVE IT.  NEVER BEEN TO HAWAII BUT IT MAKE ME FEEL LIKE I AM RIGHT THERE.  I FEEL SO CALM AFTER WATCHING THE VIDEO.  THANKS FOR INTRODUCING ME TO DREAMS OF HAWAII.       
SANDRA , TEXAS

I WANTED TO WRITE YOU AND LET YOU KNOW HOW MUCH YOUR VIDEO DREAMS OF HAWAII AS HELPED ME COPE WITH THE STRESS OF MY EVERYDAY LIFE.  WITH MY SIX CHILDREN AND MY HUSBAND WORKING 60 HOURS WEEKS WE LOOK FORWARD TO WATCHING THE VIDEO AT LEAST 2-3 TIMES A WEEK.  SOMETIMES WE JUST WATCH THE VOLCANO PART OUR FAVORITE.
THE CHILDREN LOVE IT AND PUT ON THEIR OWN MUSIC TO IT AND TRY TO DO THE HULA.  THANKS FOR BRINGING HAWAII INTO OUR HOME.  WE SURE HOPE THAT WE GET TO GO THERE SOME DAY.    BARBRA,  NEW YORK

JUST A NOTE TO LET YOU KNOW THAT MY DOCTOR HAS VERIFIED THAT MY BLOOD PRESSURE IS DOWN AND BETTER THAN IT HAS BEEN IN 10 YEARS.
I FEEL MORE RELAXED THAN EVER.
I GAVE MY DOCTOR MY VIDEO TO WATCH.  HE LIKED IT SO MUCH HE BOUGHT ONE AND PLAYS IT IN HIS OFFICE ALL THE TIME WHILE YOU WAIT TO SEE HIM.    	
STEVE ,  COLORADO

THANKS ALI?I PRODUCTIONS.  I WANTED TO WRITE YOU AND LET YOU KNOW HOW HAPPY I AM THAT I GOT YOUR VIDEO, DREAMS OF HAWAII.  IT IS HARD FOR ME TO BELIEVE THAT SUCH A SIMPLE THING AS SITTING DOWN AND WATCHING A VIDEO CAN HAVE SUCH A IMPACT ON MY LIFE BUT IT IS TRUE.
WHEN MY BOSS WANTS TO KNOW WHY I AM SO HAPPY AND UP I TOLD HIM THAT I GOT A LITTLE BIT OF HAWAII IN ME EVERY DAY.
I WATCH THE VIDEO NOW TO MY FAVORITES SONGS.  LOVE IT
BY THE WAY MY INCOME IS UP IN THE LAST 90 DAYS 70%  AND I HAVE LOST 30POUNDS.  I AM TRULY FOCUS ON BEING THE BEST I CAN BE.  THANKS 
I WILL BE COMING TO HAWAII AND HOPE TO SEE YOU ALL THERE.
DO YOU OFFER ANY OTHER VIDEO?     
EARNIE,  SEATTLE

GET YOUR OWN COPY OF DREAMS OF HAWAII TODAY AND BRING ALL THE BEAUTY AND SERENITY INTO YOUR HOME FOR YOU AND YOUR FAMILY AND FRIENDS TO ENJOY FOR YEARS AND YEARS TO COME.

SPECIAL ONE TIME OFFER

IN CELEBRATION OF WINNING THE PRESTIGIOUS GOLD TELLY AWARD IN THE FIELD OF HEALTH AND MEDICINE, WHICH IS THE FIRST TIME THAT THE GOLD TELL HAS EVER BEEN AWARDED IN THE FIELD OF HEALTH AND MEDICINE, ALI?L PRODUCTION ALONG WITH MAHALO ENTERPRISES, WILL BE OFFERING FOR THE FIRST TIME EVER A SPECIAL ONE TIME PROMOTION OF THE DREAMS OF HAWAII VIDEO AT A 25% DISCOUNT WITH AN ADDED BONUS OF A FREE VIDEO OF YOUR CHOICE. 
I URGE YOU TO TAKE ADVANTAGE OF THIS $15.00 SAVINGS TODAY. THIS IS A ONE TIME EM SPECIAL PROMOTION FOR YOU AND YOUR FAMILY.

THE GOAL AND MISSION STATEMENT AT ALI PRODUCTIONS IS TO HELP AS MANY PEOPLE AS WE CAN WITH THE REDUCTION OF STRESS.  WE KNOW THAT THERE ARE MILLIONS OF PEOPLE OUT THERE WORKING THEIR BRAINS OUT TO KEEP UP AND TRY TO GET AHEAD.  WE KNOW WE CAN HELP YOU.  
THIS SPECIAL OFFER IS FOR YOU.  PLEASE TELL A FRIEND 


When you're ready to order, use this special order form 
to get your 25% savings and free limited-edition video: 
http://www.video-hawaii.com/special_order.html

MAHALO FOR YOUR TIME,  HOPE TO SEE YOU IN HAWAII

RHONDA BORNTRAGER,  PRESIDENT, MAHALO ENTERPRISES
STEVE WOLSHIN,  MARKETING MANAGER, MAHALO ENTERPRISES
D. ROBERT BLOOM, PRESIDENT, ALI'I VIDEO PRODUCTIONS





From kernr at mail.ncifcrf.gov  Fri Aug 20 17:18:07 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Fri, 20 Aug 1999 21:18:07 GMT
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7rpksg.e6c.wtanksle@dolphin.openprojects.net>
Message-ID: <37bdc502.22161652@news.erols.com>

On Fri, 20 Aug 1999 04:03:30 GMT, wtanksle at dolphin.openprojects.net
(William Tanksley) wrote:

>On 19 Aug 1999 08:21:55 -0700, Tom Christiansen wrote:

[snip]

>>reload(module)                  First delete from %INC, then require the
>>                                module again, as in:
>>                                    delete $INC{"Module.pm"}
>>                                    require Module;
>>                                Note that Python doesn't directly support
>>                                nested modules as in
>>                                    require Dir1::Dir2::Dir3::Module;
>>                                except via a disgustingly named hackaround
>>                                module.
>
>True for 1.4, not true after that.
>
>And "ni!" isn't disgusting -- what's disgusting is what the times have
>come to, that ruffians can walk about saying "ni!" to innocent old ladies.

Besides, it could be much, much worse.  It could have been named
EckyEckyEckyEckyPikangZoopBoingGoodemZooOwliZhiv.

:-)

[snip]

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From news at dorb.com  Tue Aug  3 10:06:06 1999
From: news at dorb.com (Darrell)
Date: Tue, 3 Aug 1999 10:06:06 -0400
Subject: Annoying problem with rexec
References: <xzn1w99xwj.fsf@prakinf.tu-ilmenau.de>
Message-ID: <HHCp3.1533$ib.52332@typhoon.nyroc.rr.com>

I'll take a stab at the answer here.
You are assigning to the variable named "open" later in the function.
These variables are allocated on entry to this function.
The variable open hasn't been assigned to yet but the name
is reserved for later.

>>> def x():
...     fp=open('test')
...     open=None
...
>>> x()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 2, in x
NameError: open
>>>
--
--Darrell
Ralf Doering <ralf.doering at prakinf.tu-ilmenau.de> wrote in message
news:xzn1w99xwj.fsf at prakinf.tu-ilmenau.de...
>
> Hello,
> maybe someone can explain what happens here. Like stated in the
> subject, the problem is about the rexec module. I tracked it down to
> this simple example:
>
> def testrexec():
>     ml = ['test.1','test.2','test.3']
>     for mod in ml:
>         temp = "/tmp/"+mod
>         print "**** About to open:", temp
>         f2 = open(temp,"w")
>         f2.write("I WAS HERE.\n")
>         f2.close()
>         print "****",temp, "closed ."
>         saved_open = open
>
>         import rexec
>         r = rexec.RExec(verbose=1)
>         open = r.r_open
>         #do some stuff here like load_modules ...
>         pass
>         open = saved_open
>
> As a result, I get:
>
> >>> testrexec.testrexec_ok()
> **** About to open: /tmp/test.1
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "testrexec.py", line 8, in testrexec
>     f2 = open(temp,"w")
> NameError: open
> >>>
>
>
> I do not understand, why open becomes undefined, even *before* rexec
> is imported and before I try to do nasty things with the function
> reference. I did expect, that the first call to "open" in write mode
> should succeed, even if something goes wrong with the references to
> open later. But it seems, that I`m  wrong.
> However, if I insert a "global open" right after the start of the
> function, it works as expected.
> Anyone out there how can enlight me?
>
> Ralf
> --
> No sig -- no fun





From mhagger at blizzard.harvard.edu  Fri Aug 20 18:28:37 1999
From: mhagger at blizzard.harvard.edu (Michael Haggerty)
Date: 20 Aug 1999 18:28:37 -0400
Subject: Anyone know about lgrind?
References: <y0jyaf62o1t.fsf@vier.idi.ntnu.no>
Message-ID: <xj0ogg2cciy.fsf@cyclone.harvard.edu>

mlh at idt.ntnu.no (Magnus L. Hetland) writes:

> Or... does anyone have some generic code for marking up Python?

enscript (available from ftp.gnu.org) can pretty-print Python.  You
use a `-E python' option.

Michael

-- 
Michael Haggerty
mhagger at blizzard.harvard.edu



From fdrake at acm.org  Thu Aug 19 12:00:15 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 19 Aug 1999 16:00:15 GMT
Subject: Kudos to documentors: (popen2)
In-Reply-To: <37BC21D9.E19DBEA2@darmstadt.gmd.de>
References: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de>
	<37bc15c2@cs.colorado.edu>
	<37BC21D9.E19DBEA2@darmstadt.gmd.de>
Message-ID: <14268.10767.95070.208607@weyr.cnri.reston.va.us>

Robb Shecter writes:
 > - it should take a look at what Javadoc and Python's doc system do
 > right, and do it better.  Ie: starting with semantic markup, not just
 > formatting markup.

Robb,
  If by "Python's doc system" you are referring to the markup/tools
used for the standard Python documentation, I'd like to point out that 
although the markup language is implemented in LaTeX, it is based on
language semantics and not on presentation.  There are glitches, and
some islands of old markup remain, but those are being rooted out and
brought up to date.
  And I'm always willing to entertain suggestions for improvements!
Just send mail to python-docs at python.org to report bugs, or join the
documentation SIG (see http://www.python.org/sigs/doc-sig/) to discuss 
structural improvements.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From davidopp at megsinet.net  Mon Aug 30 02:00:48 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Mon, 30 Aug 1999 02:00:48 -0400
Subject: Upper limit?!
Message-ID: <37CA1E0F.FA2A05A3@megsinet.net>

I was playing around with the command line and discovered that when I
enter 2 to the 30th power, I get a number back...when I enter 2 to the
31st power I get an overflow error.  Is 2 to the 30th power near the
upper limit for computations?!
See below:

>>> 2**30
1073741824
>>> 2**31
Traceback (innermost last):
  File "<stdin>", line 1, in ?
OverflowError: integer pow()

David O.

"Walk this way please....."




From jbauer at rubic.com  Thu Aug 26 09:27:43 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Thu, 26 Aug 1999 13:27:43 GMT
Subject: emacs with python mode
References: <37C50DAD.840B67C5@xsoft.co.at>
Message-ID: <37C540CF.9CE369EE@rubic.com>

"Gerhard W. Gruber" wrote:
> I'm using Xemacs 21.0 Beta (August 1998) on Win32 and it works quite
> fine (is there a newer version out there?). Now my real question. When
> I'm using Python emacs recognices the mode but it doesn't automatically
> activate syntax highlighting, 

Gerhard,

I'm not very knowledgable about emacs, but I'm using
XEmacs on NT for Python editing.  Below are some settings
in my home .emacs that you might find useful.

I also find it useful to swap my CONTROL and CAPLOCKS
keys for Emacs editing on Windows machines.  Check out 
Gordon Chaffee's home page for his swapkeys utility:

  http://bmrc.berkeley.edu/people/chaffee/

-Jeff Bauer
---

;; Jeff - add ~/elisp to my load-path
(setq load-path (cons "d:/jbauer/elisp/" load-path))

;; Jeff - almost everything below this point is stuff I've modified
(setq-default indent-tabs-mode nil)
(setq make-backup-files nil)
(setq line-number-mode t)
(setq default-tab-width 4)
(setq-default case-replace nil)  ; never modify case when replacing
(setq scroll-step 2)

;; highlight the region between point and mark
(setq transient-mark-mode t)

;; Python file associations
(add-hook 'python-mode-hook 'turn-on-font-lock)
(setq auto-mode-alist
      (cons '("\\.py$" . python-mode) auto-mode-alist))
(setq interpreter-mode-alist
      (cons '("python" . python-mode)
	    interpreter-mode-alist))

;; Autoload python
(autoload 'python-mode "python-mode" "Python editing mode." t)

(cond ((fboundp 'global-font-lock-mode)
       ;; Turn on font-lock in all modes that support it
       (global-font-lock-mode t)
       ;; Maximum colors
       (setq font-lock-maximum-decoration t)))



From bowman at montana.com  Sun Aug 15 15:07:50 1999
From: bowman at montana.com (bowman)
Date: Sun, 15 Aug 1999 13:07:50 -0600
Subject: Converting a string into an object instance
Message-ID: <37B71006.48F588A1@montana.com>

Given a string representation of an object, like you would get from
id(), is there anyway to get back the actual object? 

As background, in Tkinter, it is possible to create a menu, add
cascades, and then set the 'menu' attribute of a Toplevel object to that
menu, producing a menubar. If you then create another menu containing
the pulldown entries, you can attach this menu to the toplevel cascades
with

   topMenu.entryconfigure(index, menu=pullDown)

So far, so good. However, if you then want to walk the menu tree, 
	
   topMenu.entrycget(index, 'menu') 

will return a string such as .135753368.135845912, and not a copy of the
menu object.  This is indeed the same thing you get if you 'print
pullDown', but is not too useful for manipulating the attributes or
entries of the menu.

It is easy enough to just refer to the menu directly, but in some
instances it might be useful to have the ability to modify the 4th entry
in the 3rd cascade with only a reference to the top level object.

tia

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From cbbrowne at news.brownes.org  Mon Aug 30 01:16:29 1999
From: cbbrowne at news.brownes.org (Christopher B. Browne)
Date: Mon, 30 Aug 1999 05:16:29 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <MPG.1233dfb6145766d998968c@news.mop.no> <14281.62669.466599.913229@dolphin.mojam.com>
Message-ID: <slrn7sk4t7.b2r.cbbrowne@godel.brownes.org>

On Sun, 29 Aug 1999 22:09:50 -0500 (CDT), Skip Montanaro <skip at mojam.com>
posted:
>    Alexander> The problem as I see it is that the descriptive language is
>    Alexander> not the primary tool: once you've generated your GUI source
>    Alexander> code (eg., in Python), you're separated from the original
>    Alexander> script. There's no way to re-generate the GUI framework and
>    Alexander> preserving your logic -- because there is no separation here
>    Alexander> between presentation and code. You assume they're be
>    Alexander> intermingled.
>
>It's doable.  OMTool (Jim Rumbaugh's work at GE CRD before leaving to join
>Rational) was perfectly happy to regenerate C/C++ code skeletons the user
>had already modified and get the guts put back where they belonged.
>
>One solution here would be to have two files, one that's a GUI description
>and one that's an implementation file for one of the N backend languages.
>The GUI description would indicate the name of the callback function and
>generate code to call it.  The definition of the callback would be in an
>entirely different file.  

>The user should never need to edit the generated code.
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

That is the *crucial* key to the use of these sorts of automated code
generation tools.  There *MUST NOT* be any good reason to directly edit
the code that is generated, so that it may be treated as a mere byproduct
that may be regenerated upon need.

The situation represents one where it would be Highly Desirable to use
some sort of "stub generator," ideally one that is largely
language-independent.

ILU is probably the most-developed example of such, as well as just about
any CORBA implementation.  g-wrap is a Guile-oriented scheme not unlike
this that is intended to allow declarations of interlinkages between
Scheme and C.

In the case of CORBA, there's an additional cost of having to have
requests head out to the ORB and back, which may be an unacceptable cost.

In any case, whichever sort of "stubber" might be used, it is very
important to have a clear separation between "code that needs to be
written," and "code that is automagically generated."
-- 
When replying, it is often possible to cleverly edit the original
message in such a way as to subtly alter its meaning or tone to your
advantage while appearing that you are taking pains to preserve the
author's intent.  As a bonus, it will seem that your superior
intellect is cutting through all the excess verbiage to the very heart
of the matter.
-- from the Symbolics Guidelines for Sending Mail
cbbrowne at ntlug.org- <http://www.ntlug.org/~cbbrowne/corba.html>



From frehe491 at student.liu.se  Sat Aug 28 12:49:45 1999
From: frehe491 at student.liu.se (Fredrik Henbjork)
Date: Sat, 28 Aug 1999 18:49:45 +0200
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net>
Message-ID: <99082819072100.03655@o112>

On Sat, 28 Aug 1999, you wrote:
> On Sat, 28 Aug 1999, Tim Peters wrote:
> 
> > python-folk-are-the-best-buddies-psu-dues-can-buy-ly y'rs  - tim
>                                    ^^^
> There are a few options:
> 
> 1. The mailman at Python org had a glitch
> 2. The timbot made a type
> 3. There's this underground association, Python Software Union, which goes
> around breaking thumbs of people who use Python without paying the PSA.
> 
> e-mailing-cash-and-lots-of-it-ly y'rs, Z.

Let me set the facts straight: the Python Secret Underground movement does
not exist.  Every accusation that suggests that the PSU is, or was, involved
in, or associated with, the following things is completly false: the Kennedy
assassination, Richard Stallman, black UN helicopters over Montana, NSA
and Windows 2000. We also disclaim all responsibility of having designed or
manufactured the timbot. Thank you for listening to this nonexisting broadcast.

We-don't-exist-and-we-are-not-coming-to-get-you-tomorrow-ly y'rs
Fredrik Henbjork



From jlj at cfdrc.com  Thu Aug 26 11:04:52 1999
From: jlj at cfdrc.com (Lyle Johnson)
Date: Thu, 26 Aug 1999 15:04:52 GMT
Subject: Making A Standalone Executable Under Windows NT/95/98
References: <hUQw3.117$Kl.24641667@speed.city-net.com> <007e01beef09$8d424ff0$4500a8c0@thomasnotebook>
Message-ID: <37c55740.526393784@192.77.87.74>

On Wed, 25 Aug 1999 16:53:11 +0200, "Thomas Heller"
<thomas.heller at ion-tof.com> wrote:

>[Making A Standalone Executable Under Windows NT/95/98]
>
>> Is this possible?  If so, How?
>
>Look at Gordon McMillan's win32 installer
>    http://starship.python.net/crew/gmcm/install.html.
>Additional features are included in an enhanced version of this tool at
>    http://starship.skyport.net/crew/theller/installer/.
>
>
>Thomas Heller
>

Make that last URL:

    http://starship.python.net/crew/theller/installer/
                    ^^^^^^



From eugene.leitl at lrz.uni-muenchen.de  Tue Aug 10 16:22:12 1999
From: eugene.leitl at lrz.uni-muenchen.de (Eugene Leitl)
Date: Tue, 10 Aug 1999 13:22:12 -0700 (PDT)
Subject: converting Netscrape mailbox to Unix mbox format
Message-ID: <14256.35316.328575.118034@lrz.de>

Hi, I've inherited somebody's inbox as created by Her Royal Cruftiness
Netscape. Does anybody have a Python script to salvage it back to
hallowed standards of Unix?

TIA,

Eugene



From xah at best.com  Tue Aug 24 00:29:34 1999
From: xah at best.com (Xah)
Date: Mon, 23 Aug 1999 21:29:34 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk>
	 <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu>
	 <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu>
	 <37c126a6$0$813@newsreader.alink.net>
	 <slrn7s2dtk.vsl.kaz@ashi.FootPrints.net>
Message-ID: <37c21fae$0$216@nntp1.ba.best.com>

I gather you are a newbie. Normally I'd just kick newbie's ass for the fun
of it, but this time I feel magnanimous and will give you the benefit of a
doubt.

I wrote:
>>Have you ever glanced at a graduate math text? It's probably all Greeks to
>>you. However, if someone in the know who translated all the symbolism to
>>plain English, then it becomes legible.

u wrote:
> I doubt that translating the elegant, condensed notation into verbose English
> would result in anything legible or useful.

I don't know you are in the python camp or perl camp, but apparently you are
not acquainted with the Perl Ways. English has evolved to be extremely
versatile and useful. It gives you the freedom to express anything in any
manner or style you want. If elegance AND terseness is your thing, what
better examples other than poems? (btw, you can do that in Perl too.) There
is nothing whatsoever in math symbolism that cannot be expressed better by
English. For example, a^2+b^2=c^2 can be expressed as "A number multiplied
by itself plus another number multiplied by itself is equal to a third
number multiplied by itself". Now, that's the verbose version you are
thinking of. But look, "Sum of two squared equals one" is a condensed form.
Further, "Sum two squares is one". Or if you are a power speaker: "s2is1",
which is more terse than the symbolic form and no information is lost
because of _context_. (no, pure symbolism cannot do away with context) You
know what a 'ball' is don't you? But in symbols, it's something like
a^2+b^2+c^2=d with some elaborate coordinate system and an arcana of
presumptions.

I suggest that you go to www.perl.com and read Larry's speeches of perl
conf. Anyone will do.

I wrote:
>>It really depends on who's doing the
>>reading. If you are a math guy, then perhaps symbols is your thing. Most of
>>us would prefer English. Legibility is purely subjective.

u wrote:
> If I wanted to understand the graduate math text, I'd make the effort to learn
> the symbols and whatever other background information I needed.  English would
> be needed in acquiring this information, but after that it would be a huge
> hinderance to manipulate ideas using English.

Pissss me!

Granted that a new language like math symbolism might have little
theoretical advantage down the road for some theoretical cleanness,
precision, or maybe expressiveness, but English has been used for literally
eons. The inertia is strong, the legacy is golden, and no one can stop it.
As I said before, it has also evolved into perfection. People will use
English and people will prefer English over any arcane symbolism. It is
egregious a crime that math text books are littered with arcane symbols
instead of plain English. It is why no one in the world likes math.

One thing with English is that it has not evolved into a computing language
yet, thus Perl comes to the rescue. Here's a little history lesson for you:

Larry wanted to process
some text as his job requires at Jet Propulsion Lab. However, with all
unix's power and piping shit with sed and awk and baggage but he couldn't do
it well. Therefore, following the unix tradition, he quickly hacked up his
own language and put it on the net thinking that he's done human kind a big
favor. Since unix community consists of largely of morons, they quickly
picked up Larry's tool without much decision process. Thus, it gradually
rolls into a snow ball. Perl's success is a direct result of democracy.

As a hack (by design), the early Perl is not what we know it today. It does
not have a module system, no OOP, and lacks many of the language features of
'Perl 5'. However, by the ingenious notion of evolution with Larry's
leadership, today we have all these features and beyond in the most diverse
consistency possible. Again, it does not force OOP down anyone's throat.
It's like a car without breaks pre-installed. The freedom of installation is
up to YOU.

Do me a favor and research on Perl history yourself.

I wrote:
>>However, the ivy tower class of buttheads insist on symbolism for math,
>>clinging to the fancy that certain ideas are better expressed in certain
>>language. They created a whole barrier of entry. You couldn't believe what
>>kind of perverted use of symbols they invent. I know because I've been there

u wrote:
> Mathematics is the invention and manipulation of symbols,  so you can
> hardly blame these guys for doing their jobs in life.

Mathematics is the invention and manipulation of symbols, eh??? You go back
to school!

Try to convince your professors to write damn math texts in plain English,
so you might get more bang out of your schooling.

I wrote:
>>On a personal note...  Do you know what a computer science degree is these
>>days? They require you understand shits like lambda calculus, logic, graph
>>theory, combinatorics, or even algorithms??

u wrote:
> Maybe at the ``ivy league'' schools. Where I studied I did all networking,
> advanced operating systems, compilers, advanced machine architectures,
> numerical analysis (which I flunked), that sort of thing.  With a big dose of
C
> hacking in all of these.  (Of course logic and algorithms as well, doh!)
>
> The studies that you describe are intended for people who are destined
> to become academics rather than software developers. To ``go on'' as they say.
> I think you simply went to the wrong school, one that didn't offer the hands
on
> systems programming and software development experience.

Perhaps Larry lives nearby your school and was able to make a better
curriculum by hegemony. Still, I think the logic and numerical analysis
courses should be completely dropped. Operating Systems courses should be
replaced by Unix Sys Admin. Networking courses should be replaced by TCP/IP
and DNS/Bind. (use Oreilly's excellent texts) The Compiler courses should be
replaced by teaching Perl. (we still need someone to implement Perl in Perl,
for nothing other than impressing some of the functional language asses.)

>advanced machine architectures

Is that a course teaching chips design? My favorites are Printos and
Torridos.

> Incidentally why would you be advocating Perl if you don't like terse
> ``mathematical'' notations? I don't get it. For the most part, Perl programs
> look like line noise to the uninitiated.  If you prefer English to terse
> symbols, shouldn't you be advocating AppleScript or something? Your outlook
> appears to be hopelessly inconsistent.

AppleScript only looks English, but does not have all the essential
qualities of natural languages. Like Python and others, it's too consistent
and ram some OOP or Object Models down your throat. Don't start me on Apple.
When it comes to Apple, they always push for useless change. In the past
years they degraded programers' social status with the oh--sooo-cute color
iMacs and G3 towers, and in September they'll have this
oh--sooo-cute-I'm-drooling iBook and AirPort technology probably semi-stolen
from unix's RFC standards. (RFC is an abbreviation for Really Fucking
Common)

As Larry has well said, Perl is a post-modern language. The language of the
New Age era.

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 "Perl: all unix's stupidities in one."
      -Larry Wall, from antimatter universe



From gregm at iname.com  Tue Aug 31 03:30:40 1999
From: gregm at iname.com (Greg McFarlane)
Date: Tue, 31 Aug 1999 17:30:40 +1000
Subject: My first python core dump!
Message-ID: <19990831173040.33736@nms.otc.telstra.com.au>

It may be a bug in Tk, but it causes python to core dump.

First run this:
    % python
    Python 1.5.2 (#1, Aug 31 1999, 12:54:44)  [GCC 2.9-gnupro-98r1] on sunos5
    Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
    >>> import Tkinter
    >>> root = Tkinter.Tk()
    >>> root.tk.createcommand('debugEval', eval)

Then this:
    % python
    Python 1.5.2 (#1, Aug 30 1999, 18:06:40)  [GCC 2.9-gnupro-98r1] on sunos5
    Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
    >>> import Tkinter
    >>> root = Tkinter.Tk()
    >>> root.tk.eval('send {tk #2} debugEval dir(Tkinter)')
    Traceback (innermost last):
      File "<stdin>", line 1, in ?
    TclError: target application died
    >>> 

The first python dies with:
    >>> Segmentation Fault (core dumped)

The first few lines of gdb the stack trace is:
    #0  0x57d40 in PyDict_GetItem (op=0x0, key=0x41f2f8) at dictobject.c:314
    #1  0x59bf4 in PyDict_GetItemString (v=0x0, key=0x2b2d10 "__builtins__")
	at dictobject.c:1071
    #2  0x63bb4 in builtin_eval (self=0x0, args=0x348dd0) at bltinmodule.c:732
    #3  0x700ec in call_builtin (func=0x341080, arg=0x348dd0, kw=0x0)
	at ceval.c:2359
    #4  0x6fefc in PyEval_CallObjectWithKeywords (func=0x341080, arg=0x348dd0, 
	kw=0x0) at ceval.c:2324
    #5  0xcbbc8 in PythonCmd (clientData=0x394180, interp=0x382c80, argc=2, 
	argv=0xefffde48) at ./_tkinter.c:1219
    #6  0x21bd2c in TclInvokeStringCommand (clientData=0x389880, interp=0x382c80, 
	objc=2, objv=0x3ce538) at ./../generic/tclBasic.c:1743

It looks like the call to PyEval_GetGlobals() in builtin_eval() is
returning None, which means that the current_frame is NULL, which
means I don't know.

Can anyone reproduce this?

-- 
Greg McFarlane:    INMS Telstra Australia (gregm at iname.com)
Today's forecast:  Sunny, with occasional cloudy periods and a chance
                   of precipitation in some areas.



From arcege at shore.net  Fri Aug 20 11:13:05 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 20 Aug 1999 15:13:05 GMT
Subject: Password Prompt
References: <82iu6dyned.fsf@PENELOPE.CHEM.UIC.EDU>
Message-ID: <5gev3.727$mV.257016@news.shore.net>

varner at PENELOPE.CHEM.UIC.EDU wrote:
: Hi,

: I'm writing a Tkinter interface to a program that prompts for a
: password.  I would like to pop a dialog when the program sends the
: prompt, but I'm having a little trouble figuring out how to do that.
: Right now I'm doing

: 	message = os.popen (command, 'r').readlines ()

: but the program sits at this line waiting for a response to the
: password prompt.  Any hints?

I'm not sure what ``command'' is, but it sounds like "/bin/passwd" or
the like.  Most password-prompting programs open the process's
controlling terminal (/dev/tty) and interact that way, this is for
security since stdin could be redirected from anywhere.  You will want
to use something similar to Expect to handle this (one that can handle
pseudo-terminals).  [*shameless self-plug*] One such module is
ExpectPy: http://starship.python.net/crew/arcege/ExpectPy/

  -Arcege




From philh at vision25.demon.co.uk  Sat Aug 28 20:02:23 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 29 Aug 99 00:02:23 GMT
Subject: Java project -- use Python instead?
Message-ID: <935884943snz@vision25.demon.co.uk>

I am currently involved with a largish programming project, currently
being written in Java.

It is a client-server system. The clients are PCs running NT, the
server is another NT box running a database. There will be about 5000
users with client machines (not all will be connected at a time).
The clients will connect to the server across the Internet.

The project will consist of a user interface with c. 100 screens on 
it. A Java applet displays these screens, using data it gets from the
server; the server is running Enterprise Java Beans which in turn
are connected to a relational database.

I would like to try implementing some or all of this project using 
Python, either as a prototyping system (with the ``real'' system
later implemented in Java), or as the real system.

I wonder whether anyone has any experience using Python for anything
like this? If so, perhaps you could answer some questions:

1. is it likely to be feasible to do this in Python? What about
doing it partly in Python (e.g. the client), and partly in Java
(e.g. the server)?

2. is Python likely to be fast enough? I would guess it probably is,
on the clients, but what about on the server?

3. are there any speed bottlenecks in Python, e.g. talking to the
database, or doing network communications?

4. what are the arguments for and against using Python? What's a good
argument to persuade my boss to give it a go?


-- 
Phil Hunt....philh at vision25.demon.co.uk




From mwh21 at cam.ac.uk  Fri Aug 13 02:15:25 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 13 Aug 1999 07:15:25 +0100
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk> <luemh851ym.fsf@stest.ch> <YeMs3.6620$gO1.175350@news2.giganews.com>
Message-ID: <m3u2q42ole.fsf@atrus.jesus.cam.ac.uk>

cbbrowne at news.hex.net (Christopher Browne) writes:

> On 12 Aug 1999 19:43:45 +0200, Christian Egli <christian.egli at alcatel.ch> wrote:
> >>>>>> "Phil" == Phil Hunt <philh at vision25.demon.co.uk> writes:
> >>>>>> "Paul" == Paul Prescod <paul at prescod.net> writes:
> >Paul> Glade is a GUI application that allows you to build a window
> >Paul> visually.
> > 
> >Phil> Okay, but I saw a reference somewhere to Glade having a
> >Phil> declarative file format that could be converted into code to
> >Phil> build the GUI. I was mostly talking about the declarative file
> >Phil> format.
> >
> >Maybe you meant the storage format of Glade which is XML which in
> >turn can be used with Libglade to build the GUI dynamically.
> 
> Sounds Really A Lot like CORBA DII...

Not to me. Could you elaborate for the slow of thought?
 
> >Paul> Glade does that too, as long as your "favorite language" is
> >Paul> C. :)
> >
> >Not only. See below
> ...
> >AFAIK Libglade does exactly that.
> 
> Question: How do you tie events to widgets?
> 
> Is it, maybe, perchance, via pointers to C functions???
> 

widget.connect("<signal>",<callback>).

I mean, yes, gtk+ ties events to widgets via C function pointers, but
the Python wrappers are sufficiently sophisticated that you can use a
Python function. 

pygtk rocks. I'm not sure it's as perfectly designed as could be, but
it's pretty damn cool.

HTH
Michael



From barmar at bbnplanet.com  Fri Aug 20 13:01:31 1999
From: barmar at bbnplanet.com (Barry Margolin)
Date: Fri, 20 Aug 1999 17:01:31 GMT
Subject: 'if' as a _function_, in Scheme and Python
References: <7pi0kl$ko8$1@nnrp1.deja.com> <37bc9dd6.11041486@90.0.0.40> <Ze1v3.287$m84.4462@burlma1-snr2> <37beeec3.31761662@90.0.0.40>
Message-ID: <KRfv3.310$m84.4719@burlma1-snr2>

In article <37beeec3.31761662 at 90.0.0.40>,
Steve Schafer <pandeng at telepath.com> wrote:
>On Fri, 20 Aug 1999 00:24:25 GMT, Barry Margolin
><barmar at bbnplanet.com> wrote:
>
>>No.  Remember that then-branch and else-branch are thunks.  It doesn't
>>matter whether you evaluate them, they don't do anything until you *call*
>>them.
>
>Yes, but that doesn't negate what I said--it's not equivalent to IF
>(since IF doesn't require that you wrap the expressions in thunks). 

It's not syntactically equivalent, it's another way of accomplishing the
same thing.

>Essentially, you're changing the problem to fit your solution. I guess
>I don't really see any point in doing that.

This is a discussion about language design: does "if" need to be a special
form, or can it be constructed out of other, existing primitives?  The
answer is that it can, with the only special form required being "lambda".
Yes, it means that it has to be called a little differently in the
theoretical variant of Scheme that doesn't have an "if" primitive; this can
be hidden using a macro, though.

A note about the original poster's implementation, though.  IIRC, it used
"assq" to select the appropriate thunk to call.  Can "assq" be written
without using "if"?  Although "assq" is a required language primitive, it's
a pretty high-level function and would usually be expected to be
implemented in the high-level language.  My version of "if" that used the
"if-true" and "if-false" functions really doesn't depend on anything other
than "lambda" and function calling.

-- 
Barry Margolin, barmar at bbnplanet.com
GTE Internetworking, Powered by BBN, Burlington, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.



From tchrist at mox.perl.com  Fri Aug 20 09:01:09 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 20 Aug 1999 06:01:09 -0700
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer) 
References: <001601beeaef$e795f380$712d2399@tim>
Message-ID: <37bd4385@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    "Tim Peters" <tim_one at email.msn.com> writes:
:There are real barriers for native speakers of
:either language in picking up the other.

What do you think falls into this category?  What thing 
about Python makes it hard for a native speaker thereof
to learn Perl, and vice versa?

[massive closure generator deleted]

:This one is more interesting than it looks <wink>:  in the last version of
:Perl I tried it on, it leaked memory at a prodigious rate.  

Absolutely.  You've a circular reference there.  

:The reason it's
:interesting is because it's faking pretty much what Python would have to do
:to support lexical closures, 

I'm not sure why you say "faking".

:In Python and in Perl (at least in the Perl I last tried that under) that
:creates cyclic garbage:  makefac's lexicals have to contain a binding for
:fac, and fac has to point back to makefac's lexicals in order to resolve
:up-level references.  And that's (unrecoverable trash) the reason Python
:doesn't have lexical closures today.  

That's quite right.  Doesn't python have some similar issue related
to the use of bound methods?

:Whereas Python has no warnings of any kind -- it it's irked enough at me to
:complain at all, it's irked enough to kill me.  In that it's a faithful
:reflection of Guido's personality <wink>.  Seriously, I do prefer that.

perl -w is "pardon me, but are you really sure you know what you're
doing there?", it's not "die die die, you ignorant script kiddie!".

:As a native speaker of Python, -w and "use strict" decorate all my Perl, and
:in  the few Official Perl projects I've been on "NO WARNINGS!" was an
:absolute rule.  The more fatals the better.

That's how I virtually always programmed it.  Sometimes I don't bother
on little one-liners that turn Perl into Super Sed, like

    perl -i.orig -pe 's/^(\d+)/$1 + $./e' file1 file2 file3
    # add current line number to each number that occurs at
    # the front of a line, and put changes back in place, 
    # backups into file.orig

But real programs get -w and use strict always.  After all, 
just as with /* Lint Happiness */ of days of old, I know
how to temporarily rescind them if I really am convinced
that I really do know what I'm doing.

:> What does Python do at compile-time?
:It's too busy desperately scrounging for excuses to kill your program with a
:bogus SyntaxError to do anything useful <wink>.  About the only compile-time
:optimization is that function locals (in the absence of "exec" and "import
:*" statements) are mapped to consecutive indices into a contiguous runtime
:vector.

You, Tim, might someday look at the interesting tricks that the Perl
compiler and interpreter go through about storage, speed, and tacit reuse
of my() variables (lexically scoped locals).  Remember that my() in Perl
is one of Perl's four declarations, something then that the *compiler*
knows about, and can take suitable actions based up static analysis.

(And I nearly never mean the (C or byte)-code-generating backends when 
I say "compiler" in a Perl context.   I just the guy who does the parsing
in the previous paragraph.)

I remember my frutration as a flaming C advocate (you didn't really
think I started with Perl, did you? :-) getting a job at a company
that made US$75k Fortran compilers being told that Fortran would always
be faster than C because the fc compiler could know more things about
your code than the cc compiler could, and thus make better compile-time
optimization to speed run-time.  I wonder whether Python simply hasn't
matured as far as Perl has along those lines, or whether in fact there
are inherent barriers, as there are between Fortran and C.  What do you 
think?  I also notice that Python is bereft of compiler pragmata. 

:"exec" and "import *" can alter the *set* of local names, so their presence
:inhibits that optimization.  

Aren't you glad you don't have Tcl's "upscope" naughtiness?
People sometimes ask for this in Perl, to be able pop up a level
into your dynamic caller's private lexical scope?   I'd rather have
pass-by-name. :-)

:So, no matter what anyone here tells you, Perl doesn't have a monopoly on
:being cute & repulsive at the same time <0.8 wink>.

Oh, I know.  It's the new converts who are scariest.  The adage that
one should "beware the man of one book" becomes "beware the programmer
of one language" in our world, or should that really be "the zeal of
fresh converts burns annoyingly"? :-)

--tom
-- 
 Timesharing: the use of several people by the computer



From arcege at shore.net  Fri Aug 20 13:24:15 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 20 Aug 1999 17:24:15 GMT
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu> <37BCA74C.D3DCF5DD@fnmail.com>
Message-ID: <3bgv3.759$mV.257016@news.shore.net>

Per Kistler <kistler at fnmail.com> wrote:
: Tom Christiansen wrote:

:> The fact that they did icky windows help and not man pages says
:> something significant.

: Why not create something like perldoc for python?
: There must be others missing it...

This is a big "not me".

Missing a manpage?  At times, but I've got the info files, the O'Reilly
quick reference, local copy of the HTML docs (thru lynx or netscape),
etc.

  -Arcege




From aahz at netcom.com  Tue Aug 17 15:46:16 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 17 Aug 1999 19:46:16 GMT
Subject: Help with Threads and httplib please
References: <37B9A587.3160428E@platonix.com>
Message-ID: <7pce68$bsk@dfw-ixnews6.ix.netcom.com>

In article <37B9A587.3160428E at platonix.com>,
Yishai Beeri  <yishai at platonix.com> wrote:
>
>I have a weird problem running several python threads that use httplib
>to get some pages from the web. all runs ok, until suddenly all the
>threads get stuck. A ctrl-c makes them all continue, until the next time
>they're stuck again.
>
>From what I saw in the debugger, it seems that one thread is stuck in
>the "connect" of the socket under httplib, and all the other threads are
>waiting for some reason.

We're having similar problems in a slightly different context.  I don't
have a complete answer yet, but part of the problem has to do with not
setting up the socket with a timeout.  Another part seems to be with
using blocking mode.

One thing we've definitely noticed is that if the server terminates the
socket *after* starting to send information but *before* all info has
been transmitted, the client will hang.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From davidopp at megsinet.net  Sat Aug 28 10:58:35 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Sat, 28 Aug 1999 10:58:35 -0400
Subject: Suitable for Large Applications?!
Message-ID: <37C7F91A.B37535@megsinet.net>

Dear Group,

I have just started learning Python and would like to hear more
discussion about its suitability for large multithousand line
apllications.  On one hand I hear people discussing applications that
have 10,000 plus lines and on the other I hear people say its really
only good for small tasks and for modeling.  I'd hate to spend time
getting real good at this language only to realize that its not well
suited to large applications...
Do Python programs need to be converted to C or C++ and then compiled to
be really useful AND run fast?  Or can you get away with creating a
freestanding  executable by packaging the bytecode in freeze or another
program like installer?
What are the real performance issues.  When I hear someone say that
copiling the Python code took hours and compiling equivalent C code took
seconds, tends to get a newbie like me less enthused.  Should I take
back my O'Reilly Python book and get a book on learning C++?!

David O.





From mlh at idt.ntnu.no  Tue Aug 24 19:53:37 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 25 Aug 1999 01:53:37 +0200
Subject: word1.word2 notation
References: <oxgbtbxejea.fsf@panix7.panix.com>
Message-ID: <y0jzozgybum.fsf@vier.idi.ntnu.no>

Yeoh Yiu <squid at panix.com> writes:

> I've noticed in various languages, there is a construction
> of expressions with a dot in the middle.  It is a construction
> because the name with a dot in the middle is never explicitly 
> declared, as in 
> 
> 	int foo.bar = 42;   // doesn't happen this way
> 
> but rather foo is declared in one place and method (or some other 
> type of entity)  bar was declared somewhere else and foo.bar was
> always available implicitly as a latent method (or as something).
> 
> What is the name of this technique and the name of methods 
> (or variables ?) so constructed ?

Hm. Maybe I'm missing something here... But if this question is as
basic as I think, then maybe this will be an adequate answer:

The construction you are referring to is a "member reference". In the
example above, you refer to the member variable "bar" of the object
"bar". In cases in which the member is a function (of - actually, a
method) which is called in the reference, it is called a "method
call", as in:

  foo.bar()

You might want to look into how object-oriented programming works in
general to further appreciate the meaning of this.

> 
> squid.
> 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From pnpayne at swissonline.ch  Fri Aug 20 14:49:31 1999
From: pnpayne at swissonline.ch (Philip Payne)
Date: Fri, 20 Aug 1999 20:49:31 +0200
Subject: Python+Apache+Oracle+???Persistent database connection???
Message-ID: <37BDA33B.E8DFEBC5@swissonline.ch>

Hi,

Can any of you Python gurus tell me if the following would be
easy/possible/impossible and perhaps point me in the right direction???

I have Python (1.5.1), Apache (1.3.6), Oracle (8.0.5), PyApache (4.16),
and DCOracle (Python interface to Oracle - 1.2.1) on my Linux  box
(Caldera OpenLinux 2.2). It works fine.

But what I would like to have is a persistent database connection for
each Apache child server that is accessible to every Python script that
is fired up by that server. This because logging on to Oracle is VERY
expensive compared to simple database accesses. As far as I can see this
isn't possible with PyApache/DCOracle at present.

Now my Python level is middling, I have no experience in embedding or
extending Python, or in writing Apache modules, and my C is *VERY*
rusty, but from browsing some of the documentation and code this is my
limited understanding of what happens every time when Apache calls
PyApache to run a Python script:
        Py_NewInterpreter
        execute Python script
        Py_EndInterpreter

Logging on to Oracle in DCOracle returns a 'Connection' object. Am I
right in thinking that this object and every other Python object
automatically and ineluctably get destroyed by the Py_EndInterpreter
call??? If yes, is there any way to save it away 'somewhere' and get it
back later 'somehow' after the next Py_NewInterpreter call??? Or would
this require some sort of coordination between the PyApache and DCOracle
code??? Or is it impossible???

Any help gratefully appreciated.

Regards,
Philip Payne



From pduffin at mailserver.hursley.ibm.com  Fri Aug  6 05:42:17 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 06 Aug 1999 10:42:17 +0100
Subject: How to link statically with python15.lib....
References: <sYdq3.157$E%3.952@news1.online.no>
Message-ID: <37AAADF9.4B76@mailserver.hursley.ibm.com>

Egeland wrote:
> 
> Hello. Could someone please tell me how to link statically to python15.lib
> using MSVC6.0 Prof. Edit. ?
> I want to do this in order to avoid messing with *.dll's that may reside on
> the installation platform. I'm not
> really into install shield or anything ; just wanted to do it the easy way
> (if there is one). Guess I'll have to
> link to tcl80.lib and tk80.lib as well..?
> 

If you rebuild Python as a static library embedded in an application
you will not be able to dynamically load any C extensions because Python
relies on backlinking which is not supported on Windows, nor can it be
emulated on Windows as it is on the various Unixes (AIX is what I know)
which don't support it "properly".

Tcl and Tk on the other hand can be statically linked into an 
application and still dynamically load C extensions because it uses
its 'own' linking mechanism.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From mal at lemburg.com  Fri Aug 20 04:04:17 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 20 Aug 1999 10:04:17 +0200
Subject: A question of variables
References: <37bcf7d7.0@sahara.cablecat.com>
Message-ID: <37BD0C01.301F401C@lemburg.com>

Carlos Ors wrote:
> 
> How can I declare a variable as a private variable inside a method of a
> class?

All variables defined inside a method of a class are private to
that method except ones declared global with the global statement.
The same applies to functions.

"Private" instance attributes can be created using the __XXX 
convention, e.g. self.__privatevar = 1. These are only
useable for methods defined in the same class, not for methods
defined in subclasses.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   133 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From catfood at apk.net  Sun Aug 22 14:04:59 1999
From: catfood at apk.net (Mark W. Schumann)
Date: 22 Aug 1999 14:04:59 -0400
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b62939@cs.colorado.edu> <7ph3p8$7ci@junior.apk.net> <7phs0o$17q@dfw-ixnews16.ix.netcom.com>
Message-ID: <7ppe4b$1h6@junior.apk.net>

In article <7phs0o$17q at dfw-ixnews16.ix.netcom.com>,
Eric Bohlman <ebohlman at netcom.com> wrote:
>Mark W. Schumann (catfood at apk.net) wrote:
>: I've seen
>: 
>: #define ARBEGIN {
>: #define AREND   }
>: 
>: in Clipper (a somewhat C-ish language in this regard) code.
>: 
>: Eh?
>
>In Clipper, curly braces are used as anonymous array constructors, so 
>those mnemonics are presumably short for "array begin" and "array end."

Yes, I know.

What I don't know is how the heck this guy thought that made the code
more clear!




From arcege at shore.net  Tue Aug 17 10:50:37 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 17 Aug 1999 14:50:37 GMT
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org> <14265.5096.537800.149227@buffalo.fnal.gov> <37B91E9A.2451936C@w3.org> <001301bee891$82fccc60$f29b12c2@secret.pythonware.com>
Message-ID: <1Feu3.381$mV.141087@news.shore.net>

Fredrik Lundh <fredrik at pythonware.com> wrote:
: (posted and mailed, as usual.  or to be precise, mailed and mailed ;-)

: Dan Connolly <connolly at w3.org> wrote:
:> >  > I hope the os module will be deprecated in future releases.
:> > 
:> > I don't think there is much chance of this happening, it it too widely
:> > used.
:> 
:> How about attempting to change that trend? i.e. how about
:> in stead of documenting os.getpgrp() with an availability
:> caveat, the os module documentation only lists the stuff
:> that's really expected to be portable, and has a note
:> at the bottom ala:
:> 
:> the os module may contain other implementation-defined
:> objects; for example, on posix platforms, the contents
:> of the posix module are shadowed in the os module.
:> [... list the other cases ...]

: sounds reasonable.

: (in other words, it's mostly a documentation
: problem, rather than a design problem...)

: (Fred?  Guido?)

The better question is:
   Why did it change from documenting these components in the OS-
   dependant sections (specifically only posix) to documenting them
   mostly in the module-os section in the current release?

Except for "New in version 1.5.2." comments, the pre-1.5.2
documentation listed nothing more than exec[lv]* functions and there
was only the "posix" module documented.

  -Arcege




From jstevens at basho.fc.hp.com  Fri Aug 13 22:53:40 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 19:53:40 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132002.OAA03038@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:02 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cc24@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :Perl is harder to learn how to read, harder to remember how to read,
> :and harder to read if you, yourself, did not write it.
> 
> Proof my assertion, eh?  I disbelieve.  I want to see quantifiable,
> reproducible, and consequently irrefutable proof that Perl code that I
> did not write is harder for me to read than Python code I did not write.

Been there, done that.

No matter what I said, you would disbelieve.

That's Ok.  Every language needs its proponents.

I've done the test often enough, with enough different groups to
be confident in my statement (assuming that you understand that
"you" means the average programmer, not you, Tom).

John S.




From rjroy at takingcontrol.com  Wed Aug 25 09:17:20 1999
From: rjroy at takingcontrol.com (Robert Roy)
Date: Wed, 25 Aug 1999 13:17:20 GMT
Subject: xanadu and python
References: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com>
Message-ID: <37c3ec06.54169312@news1.on.sympatico.ca>

On Wed, 25 Aug 1999 08:03:19 GMT, "Fredrik Lundh"
<fredrik at pythonware.com> wrote:

>I doubt anyone have missed that Ted Nelson's Xanadu,
>the original hypertext system, has finally been released
>as open source:
>
>    http://www.udanax.com/
>
???
I epyt sdrawkcab oot


>what Python has to do with it?  well, go to the site and
>find out for yourself ;-)
>
></F>
>




From webmaster at chatbase.com  Sun Aug 15 21:43:41 1999
From: webmaster at chatbase.com (Tim Greer)
Date: Sun, 15 Aug 1999 18:43:41 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ce82@cs.colorado.edu>
Message-ID: <37B76CCD.1D7FCD82@chatbase.com>

"John W. Stevens" wrote:
> 
> > In comp.lang.perl.misc,
> >     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> > :With PCRE, I prefer Python.  I am slowly giving up Perl altogether,
> > :except as a training language in OO classes.  Not to surprisingly,
> > :Python is preferred more than 7 to 1 over Perl by students who are
> > :exposed to both at the same time.
> >
> > I'm sure I am perfectly capable of presenting both Perl and Python
> > in such a way that the students would prefer Perl by 7 to 1 as well.
> 
> Except that in your case, it would a concious attempt to influence
> the outcome.
> 
> In my case, it was not.  In fact, the first four times I did this,
> I was subconciously favoring Perl.  Imagine my shock. . .
> 
> > So what?
> 
> So. . . draw your own conclusions.  I've already seen how you
> structured one reply.
> 
> John S.

Sounds to me, John, by your statements throughout this thread, that your
"students" preferred Python over Perl, simply based on their observation
of you failing at Perl and making more sense out of Python, due to the
fact that you seem to understand Python a hell of a lot more then Perl.
If an instructor is teaching a course on Auto mechanics and makes
complete and comprehensible sense out of Chevy engines and find himself
confused when working on a BMW or a Porshe, I think I'd be incline to
favor the American engine over the foreign, at least until I found
someone knowledgeable enough, or learned on my own of some of the
advantages that some of the foreign motors offer, assuming this was even
a possible fact. bad real world example, but your statement in this
thread did little good for your argument.
-- 
Regards,
Tim Greer   : webmaster at chatbase.com  | software at linkworm.com
The ChatBase: http://www.chatbase.com | 250,000+ hits daily Worldwide!
TRG Software: http://www.linkworm.com | CGI scripts in Perl/C, & more.
Unix/NT/Novell Administration, Security, Web Design, ASP, SQL, & more.
Freelance Programming & Consulting, Musician, Martial Arts, +Sciences.



From akuchlin at mems-exchange.org  Wed Aug 25 14:56:47 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 25 Aug 1999 14:56:47 -0400
Subject: pyKDE
References: <rs2olmiv5u288@news.supernews.com> <slrn7s3obr.gh.amk@207-172-112-26.s26.tnt4.ann.va.dialup.rcn.com> <rs62httkdmc22@news.supernews.com>
Message-ID: <3dwvujwuxc.fsf@amarok.cnri.reston.va.us>

Jeremy <jeremy at exit109.com> writes:
> Further investigation reveals that libkdecmodule.so isn't loading up the
> Qt or KDE libraries, or any libraries at all.  It's not failing, it's
> just not even trying.  The link flags look okay during the compile (I
> compiled it again and played with it a little).  I haven't figured out

Try playing with different parameters to dlopen() in
Python/importdl.c, in the Python source code.  I vaguely recall
problems with RTLD_LAZY vs. RTLD_GLOBAL affecting other people.  (On
the other hand, I didn't do any such hackery when I compiled pyKDE.)

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
First things first, but not necessarily in that order.
    -- The Doctor, in John Flanagan and Andrew McCulloch's _Meglos_




From info at pythonware.com  Thu Aug 12 12:37:27 1999
From: info at pythonware.com (PythonWare)
Date: Thu, 12 Aug 1999 16:37:27 GMT
Subject: Tkinter newbie question
References: <cThs3.135$Vn5.17296@newsfeed.avtel.net> <37B228AD.A282DE30@cosc.canterbury.ac.nz>
Message-ID: <004001bee4e0$f7659730$f29b12c2@secret.pythonware.com>

Greg Ewing <greg at cosc.canterbury.ac.nz> wrote:
> There is also some alternative Tkinter documentation 
> available now, written with less reference to Tcl terms. 
> See
> 
> http://starship.python.net/crew/fredrik/tkintro/Introduction.htm
> http://starship.python.net/crew/fredrik/tkclass/

those copies are quite old (should really be
updated some day).  for the latest versions,
see:

    http://www.pythonware.com/library.htm




From greg.ewing at compaq.com  Tue Aug 17 18:05:41 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 18 Aug 1999 10:05:41 +1200
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org> <14265.5096.537800.149227@buffalo.fnal.gov>
Message-ID: <37B9DCB5.7B29ADF6@compaq.com>

Charles G Waldman wrote:
> 
> I think it's good advice.  If you're not using features like
> "getpgrp", but more mundane things like "listdir", then

If you're not using "getpgrp", then you won't miss it
from the os module.

I agree with Dan in that the os module (or future generic 
replacement) should only contain portable functions. If 
you want a platform-specific function, you should have to 
import a platform-specific module.

Greg



From cmcurtin at interhack.net  Fri Aug 27 10:17:37 1999
From: cmcurtin at interhack.net (Matt Curtin)
Date: 27 Aug 1999 10:17:37 -0400
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com> <001f01bef078$25fcf870$f29b12c2@secret.pythonware.com>
Message-ID: <xlxk8qhs3y6.fsf@gold.cis.ohio-state.edu>

>>>>> On Fri, 27 Aug 1999 10:37:22 GMT,
    "Fredrik Lundh" <fredrik at pythonware.com> said:

Guy_Oliver> Why do we call python a scripting language?

Fredrik> who's calling it that? ;-)
Fredrik> http://www.python.org/doc/Summary.html says:

[Correct definition, sans mention of "script", snipped.]

Nevertheless, in practice, people tend to view things like Perl and
Python as "scripting languages", largely because they see that you
"run the source" (i.e., have a shebang line in Unix-land or
"associate" a particular source filename extension with an
interpreter).

This is unfortunate.

I rather like Larry Wall's definition of "script": something that you
write to accomplish a quick job at hand, as opposed to a "program":
something you write with the intention of putting it into some sort of
production.  It's a little difficult for most people to get their
hands around this definition, though, because it means that you can
have things like C scripts and shell programs.

But I think that the "whether it's a script or program depends on what
you do with it" approach is correct.

-- 
Matt Curtin cmcurtin at interhack.net http://www.interhack.net/people/cmcurtin/



From MHammond at skippinet.com.au  Tue Aug  3 19:16:50 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Wed, 4 Aug 1999 09:16:50 +1000
Subject: Python in web page
References: <37a70f28.96313821@midpoint>
Message-ID: <7o7ta9$roe$1@m2.c2.telstra-mm.net.au>

Under Microsoft Internet Explorer you can use Python as an Active Scripting
language.

Mark.

Timothy Dixon wrote in message <37a70f28.96313821 at midpoint>...
>Is it possible to host a Python application in a web page?  TCL has a
>plug-in for IE and Netscape; is there something similar for Python?
>





From aahz at netcom.com  Sun Aug  8 15:57:19 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 8 Aug 1999 19:57:19 GMT
Subject: Parsing HTML tags with 're'
References: <7okkte$25a$1@nnrp1.deja.com>
Message-ID: <7oknev$kac@dfw-ixnews15.ix.netcom.com>

In article <7okkte$25a$1 at nnrp1.deja.com>,
Bruce Fletcher  <befletch at my-deja.com> wrote:
>
>As I said before, I would also be interested to know if there is a much
>better way to do this.  Did I miss a standard HTML parsing Python module?
>If this approach is reasonable, feel free to scoop the code for your own
>nefarious purposes.

sgmllib/htmllib
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From Friedrich.Dominicus at inka.de  Fri Aug 27 09:32:19 1999
From: Friedrich.Dominicus at inka.de (Friedrich Dominicus)
Date: Fri, 27 Aug 1999 15:32:19 +0200
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net> <37C552DF.9F90BA36@megsinet.net> <slrn7sblvo.p96.wtanksle@dolphin.openprojects.net>
Message-ID: <37C69363.DC8B6723@inka.de>

William Tanksley wrote:
> 
> On Thu, 26 Aug 1999 10:44:48 -0400, David Oppenheimer wrote:
> 
> >That being said, are there ways to integrate Python and Perl to do things
> >that neither one can do alone?!
> 
> Minotaur is an attempt to do exactly that -- they've got Python, Perl,
> Tcl, Ruby, Forth, and several other languages all glued together and all
> callable from each other, with good data interchange.

Would be helpful to know how they are doing and where

Regards
Friedrich



From kistler at fnmail.com  Fri Aug 20 19:28:03 1999
From: kistler at fnmail.com (Per Kistler)
Date: Sat, 21 Aug 1999 01:28:03 +0200
Subject: Setting environment variables.
References: <37BD3429.7364B4F7@iinet.net.au> <37BD3A33.6CAA145A@fnmail.com> <19990820175347.F461@xs4all.nl>
Message-ID: <37BDE483.7621AE94@fnmail.com>

Hi Thomas

I don't know UserDict yet, but it seems to work like that:

#!/per/bin/python
# Change env and mutate back to a shell

import os

shell   = "/usr/bin/bash"
os.environ["MYVAR"] = "MYVALUE"
env = {}
for k,v in os.environ.items(): env[k] = v
os.execle( shell, "-", env )   

Then I end up in a shell with the new environment:-)

Thanks a lot (, although it was not my original question)!
-Per.     

Thomas Wouters wrote:

> >>> x = UserDict.UserDict(os.environ)
> >>> x["spam"] = "eggs"
> 
> Just so you know i aint cheating this time ;-)
> 
> >>> os.environ["spam"]
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
>   File "/usr/lib/python1.5/UserDict.py", line 14, in __getitem__
>     def __getitem__(self, key): return self.data[key]
> KeyError: spam
> 
> >>> os.execle("/bin/tcsh", "-", x)
> centurion:~/python > echo $spam
> eggs
> 
> Yes, you could set os.environ before an os.exec*(), but passing a modified
> env is nicer.
> 
> --
> Thomas Wouters <thomas at xs4all.net>
> 
> Hi! I'm a .signature virus! copy me into your .signature file to help me spread!

-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From see-sig at wavedivision.com  Mon Aug  2 16:06:18 1999
From: see-sig at wavedivision.com (Rick Robino)
Date: Mon, 02 Aug 1999 20:06:18 GMT
Subject: hidden fields in HTML forms
Message-ID: <_Smp3.1$_j3.664@smithers.wenet.net>

Greetings,

Could anyone point me in the right direction of finding a cgi class
that catches hidden fields sent to it from an HTML form? cgi.py doesn't
recognize them; HTMLgen does define an input class which includes
hidden fields which it can create but that won't help much in parsing
already created form output except for deriving my own class.

Being lazy, and new to python, I thought I would check to see if there
was already a hidden-type class or something that I missed among
common modules. In trying to learn python, my first project has been
to create a very generic cgi mailform (like the popular perl one), and
hidden fields would be very helpful in abstracting the cgi details.

Any tips, criticism, pointers, advice are all welcome. Oh yeah, I do
know about Zope but I'm trying to make this a standalone improvement
on the perl version of mailform.

TIA,

--Rick 
                                ^^^^^^^^^^^^^^
Rick Robino                                  mailto:rrobino at wavedivision.com



From ron_savage at non-hp-australia-om5.om.hp.com  Sun Aug  1 22:53:23 1999
From: ron_savage at non-hp-australia-om5.om.hp.com (Ron Savage)
Date: Mon, 2 Aug 1999 12:53:23 +1000
Subject: Q: How to use sio-151.zip
Message-ID: <7o31l7$1ij$1@ocean.cup.hp.com>

Folks,

I'm a Perl programmer, just starting Python. I need a Python serial port
lib, so I downloaded sio-151.zip.

I'll be working under bash under WinNT.

The question is, what to do next?

I've been reading http://www.python.org/doc/FAQ.html, but it's not clear
what needs to be done to install a Python lib (if indeed sio needs to have
anything done to it to install it), or if I just point PYTHONPATH at the dir
containing serial.pyc.

--
Cheers
Bus: rons at hpaco.aus.hp.com
Home: ron at savage.net.au
http://savage.net.au/





From pehr at alum.mit.edu  Thu Aug  5 23:24:17 1999
From: pehr at alum.mit.edu (pehr anderson)
Date: Thu, 05 Aug 1999 23:24:17 -0400
Subject: Oneliners?
References: <378D99C4.BDACE15B@serop.abb.se> <004601beceb4$15711440$f29b12c2@secret.pythonware.com>
Message-ID: <37AA5561.A51E2B44@alum.mit.edu>

Dear Fredrik,
I am doing some python UI work
and am in *desparate* need of a good reference
on python/tkinter. Would it be possible for me
to gain access to a preprint of what you have
finished so far on your book?

I already have a copy on order with fatbrain,
but as usually, all projects must be done in "internet" time.

I work at E Ink corporation, developing a new kind of
electronic display, and of course I use python wherever
possible in building UI and utilities for entering in information
to be displayed in our signs.

I'd be willing to send you my feedback, and I'd be eternally greatful.

    Thank in advance,
    pehr anderson

    www.nbxcorp.com '96
    MIT '97
    www.media.mit.edu '98
    www.eink.com '99


Fredrik Lundh wrote:

> Thomas S. Strinnhed <thstr at serop.abb.se> wrote:
> > I heard there was a webserver in one line of Perl,
> > does Python match that (or two or three...)?
>
> how about:
>
>     from asyncHTTPServer import *; HTTPServer(80, FileSource("/web")); mainloop()
>
> (one line, three statements)
>
> writing the asyncHTTPServer module is left as an exercise
> (or you can wait for my Python/Tkinter book...)
>
> </F>




From chinaski_py at my-deja.com  Thu Aug 26 01:55:38 1999
From: chinaski_py at my-deja.com (chinaski_py at my-deja.com)
Date: Thu, 26 Aug 1999 05:55:38 GMT
Subject: MSDOS window
References: <7q2brn$u7h$1@nnrp1.deja.com>
Message-ID: <7q2ksn$41i$1@nnrp1.deja.com>

In article <7q2brn$u7h$1 at nnrp1.deja.com>,
  henry299 at my-deja.com wrote:
> Hi, i'm new at this. I just downloaded version
> 1.5.2 for windows 95. When i double click on a
> program in the windows explorer, (example.py
> which is  print "This is some example text") an
> MS-DOS window opens and the text is displayed,
> but the window closes by itself in less than a
> second. Am i doing something wrong, and is there
> a way to keep the popup window open? Any help
> would be greatly appreciated.
>
> Thanks in advance,
>
> Matt
                          I'm a newbie too but It's not you ,it just
does that.It opens,executes and exits.If you want to use the DOS
prompt,you need to import the file.When the prompt opens, just type:
                    import example (or other filename)
Don't put the .py at the end of example, it will error and tell you that
there is no module(file) named py.Another way to open it is through IDLE
or PythonWin(having Windows also, this is a little more familiar
ground)If you don't have PythonWin, get it at Python.org(Win32all
build125).If your new to programming or just Python, feel free to email
me at   strat_addict at yahoo.com   A few other folks and I (all newbies)
correspond to each other to help each other out.
                                 Good Luck,
                                       G.Norton






>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bwarsaw at cnri.reston.va.us  Thu Aug 19 14:39:35 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 14:39:35 -0400 (EDT)
Subject: Kudos to documentors: (popen2)
References: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de>
	<37bc15c2@cs.colorado.edu>
	<37BC21D9.E19DBEA2@darmstadt.gmd.de>
	<14268.10767.95070.208607@weyr.cnri.reston.va.us>
	<37BC327A.3DA151EA@darmstadt.gmd.de>
Message-ID: <14268.20327.730594.477713@anthem.cnri.reston.va.us>

>>>>> "RS" == Robb Shecter <shecter at darmstadt.gmd.de> writes:

    RS> What I eventually did was start a Python interpreter, do a
    RS> method call, and print the returned objects - which told me
    RS> their types.

And this actually points out one of Python's greatest assets, its
interactive interpreter.  It's even cooler in JPython where you can
just import a random Java class and fiddle with it to see what it
does.

-Barry



From kas at maps.magnetic-ink.dk  Tue Aug  3 14:11:28 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Tue, 3 Aug 1999 20:11:28 +0200
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <7o78fb$blb$2@chief.death.de>
Message-ID: <slrn7qec6g.oet.kas@titan.magnetic-ink.dk>

Felix Puetsch wrote:

>> <BODY>
>> {
>>   print n*3
>> }
>> </BODY>
> 
> How would you make a web page display this message, e.g.? I think you 
> should at least define environments for Python code. ('<!--' etc.)

I think that's what Ian tried to imply with the {} pair, I didn't perceive
his article as an implementaion, rather as an idea to be elaborated on.


  //Klaus

-- 
???[ Magnetic Ink ]????????????????????????????????????????????????? ><> ???
???[ Key fingerprint: 2E37 EAED F848 815B 050A FBAA 8F68 DD66 B4B3 7EC4 ]???



From mal at lemburg.com  Fri Aug  6 05:08:32 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 06 Aug 1999 11:08:32 +0200
Subject: Get 'ifconfig' information via Python
References: <87yafs8ruu.fsf@linuxcare.com> <37A81400.7474FD89@lemburg.com> <37A82B77.FB3D03E2@callware.com>
Message-ID: <37AAA610.6DC224EB@lemburg.com>

Ivan Van Laningham wrote:
> 
> > The fcntl.ioctl() function can take a string or an integer as
> > third argument.
> 
> [snip]
> 
> >
> > Building these structs can be done using the Python struct
> > module... playing around with this can probably crash your system
> > though.
> >
> > In the end, I think you're better off hacking together a
> > new extension module.
> >
> 
> Why on earth would reading this information into a struct crash his
> system?  Or am I missing something here?  At worst, writing a messed-up
> struct might hose his ethernet connection, but I really doubt that he
> would be any more likely to crash his system using Python than C.

The ioctl() gives you access to very low-level structures of
the whole system, not only the ethernet devices. Started as root
(which he would have to do to set the ethernet hardware address),
such a script could do plenty of damage in all kinds of weird areas
that are accessible to ioctl().

In David's case only his ethernet device would go wild, I guess.
I still think that accessing the data via an extension is safer
-- at least the module won't work on platforms that the extension
was not compiled for. The Python version will run on all platforms
having fcntl.ioctl() and it is likely that the device struct
layout changes from platform to platform.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   147 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From johann at physics.berkeley.edu  Thu Aug 19 00:50:33 1999
From: johann at physics.berkeley.edu (Johann Hibschman)
Date: 18 Aug 1999 21:50:33 -0700
Subject: Deepcopying numeric array
References: <37BA93CB.5050@cs.bham.ac.uk> <37BB8989.7BA08CA3@compaq.com>
Message-ID: <mtogg45q7a.fsf@astron.berkeley.edu>

Greg Ewing writes:

> Stuart Reynolds wrote:
>> 
>> Does anyone know why the Numeric arrays can't be deepcopied? Is it just
>> an omission?

> Not only that, but I just found out a few minutes
> ago that they're un-comparable and un-hashable!

> Is there any good reason for these restrictions?
> Are they on hold awaiting Rich Comparisons or
> something?

Well, they're mutable, so you can't use them as hash keys anyway, and,
yes, comparisons are Broken By Design since they can't "do the right
thing."

-- 
Johann Hibschman                           johann at physics.berkeley.edu



From philh at vision25.demon.co.uk  Tue Aug 31 18:52:47 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 31 Aug 99 22:52:47 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <935457373snz@vision25.demon.co.uk> <gr6gq7.jd.ln@bigbird.sw.asiainfo.com>
Message-ID: <936139967snz@vision25.demon.co.uk>

In article <gr6gq7.jd.ln at bigbird.sw.asiainfo.com>
           markh at usai.asiainfo.com  writes:
> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> : In article <7ps4ma$n06$1 at wagner.wagner.home>
> :            vitus at wagner.rinet.ru "Victor Wagner" writes:
> 
> :> Really it is very simple to define parrot interpreter on tcl - just
> :> define commands such as window, columnLayout, rowLayout as tcl procedures 
> :> which substitute
> :> something by default.
> 
> : Would anyone want to do this in Tcl as opposed to Python? (Not
> : wishing to get into another language war :-)).
> 
> Interestingly, your parrot specifications follow Tcl's syntax,

Great minds think alike, eh? :-)

BTW, I might have a bare bones implementation of the tkinter
backend up by this weekend, if all goes well.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From skip at mojam.com  Fri Aug 27 19:54:18 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 27 Aug 1999 18:54:18 -0500 (CDT)
Subject: Fuzzy string matching?
In-Reply-To: <19990827231353095.AAB161.428@max41051.izone.net.au>
References: <012501bef061$10db8b30$f29b12c2@secret.pythonware.com>
	<19990827231353095.AAB161.428@max41051.izone.net.au>
Message-ID: <14279.9340.845977.120321@dolphin.mojam.com>

    John> Here are some illustrative results from the Python-supplied soundex 
    John> module, which generates a 6-byte key:

	...

    John> This module is incorrect in its handling of multiple occurrences of the 
    John> same code:

	...

I am guilty of dropping the ball.  Guido would like sondex.c out of the
core.  Tim Peters sent a Python replacement to me.  I was supposed to give
it some life.  It's still sitting in my inbox.  I'll try to take care of it
next week (bug fixes and all).

(I realize that doesn't make Soundex any better for the fuzzy matching
algorithm, but I should at least fix any obvious bugs...)

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From nriedlin at my-deja.com  Wed Aug 18 04:31:45 1999
From: nriedlin at my-deja.com (nriedlin at my-deja.com)
Date: Wed, 18 Aug 1999 08:31:45 GMT
Subject: Controlling Python
References: <7p93ge$2bt$1@nnrp1.deja.com> <37B82DDA.2DE0E0A6@lemburg.com>
Message-ID: <7pdr1h$gto$1@nnrp1.deja.com>

> You can use a callback which you can register with the
> sys.settrace() function. A less fine-grained hook is available
> through sys.setprofile(). See the docs for more infos.
>

You are correct, but isn't it true, that the callback is only called
every function call? I would like to interfere every line (and so being
able to trace my scripts). Please correct me if I'm wrong.

TIA

Norbert Riedlin


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bogw at geo.vu.nl  Fri Aug 20 05:20:00 1999
From: bogw at geo.vu.nl (Patrick Bogaart)
Date: Fri, 20 Aug 1999 11:20:00 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <slrn7ro8n1.h6v.jblaine@shell2.shore.net>
Message-ID: <37BD1DC0.C6F07B08@geo.vu.nl>

Jeff Blaine wrote:
> 
> >GOTCHA: (low)
> >    You can't use "in" on dicts.  Instead, you must use
> >       d = { "fred":"wilma", "barney":"betty" }
> >       if d.has_key("fred"):   # legal
> >       if "fred" in d:         # ILLEGAL
> >    I don't understand why this was done.
> 
> I'm very curious what would you think should be legal.
> 
> Would 'in' (when used on a dict) search the keys, values, or both?
> 
> Each one of those seem like a bad idea to me.  You have 2 separate
> areas containing data.  Searching each of those individually
> (has_key()...) seems right to me.

If you need to search *both* data areas,
why not do it the obvious way, then?:

	d = { "fred":"wilma", "barney":"betty" }
	if "fred"  in dict.keys():   # legal, and true
	if "wilma" in dict.values(): # also legal, also true

logic: dictionaries do have a "has_key()" methods, but no "has_value()"
method
drawback: keys() and values() create copies of the dict's list of keys,
values,
so has_key() is much faster..

--
Patrick W. Bogaart
Department of Geomorphology and Quaternary Geology
Faculty  of Earth Sciences,  Vrije Universiteit Amsterdam
bogw at geo.vu.nl            http://www.geo.vu.nl/users/bogw



From frankn=news at cs.vu.nl  Fri Aug  6 08:32:27 1999
From: frankn=news at cs.vu.nl (Frank Niessink)
Date: Fri, 6 Aug 1999 12:32:27 GMT
Subject: Help with dictionary, please
References: <7oek07$so3$1@alexander.INS.CWRU.Edu>
Message-ID: <7oekkr$b43@cs.vu.nl>

rbl at hal.cwru.edu wrote:
> 	if prev1 != '' and prev2 != '':
> 	 	   key = (prev2,prev1)
> 	           dict[key] = dict[key] + 1
>    and got this error:
> 	  File "Metro:Python 1.5.1:learn2.py", line 34
>           dict[key] = dict[key] + 1
>                          ^
> 	SyntaxError: invalid token

You need to initialize the value at dict[key], it is not automatically
set to 0 or something like that. For example:

	if dict.has_key(key):
		dict[key] = dict[key] + 1
	else:
		dict[key] = 1

However, you should've got a KeyError, I don't understand why you're
getting the SyntaxError...  

HTH, Frank

-- 
He went across to the telephone machine and fiddled and fumed with all its
buttons for a while, because it was the one which was particularly recom-
mended by Which? magazine and is almost impossible to use without going mad.
	-- Douglas Adams, 'So long, and thanks for all the fish'



From chad at vision.arc.nasa.gov  Sun Aug 29 15:39:40 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Sun, 29 Aug 1999 19:39:40 +0000
Subject: Why use Perl when we've got Python?!
References: <000801bef12c$fef6e140$ad2d153f@tim>
Message-ID: <37C98C7C.724A58CB@vision.arc.nasa.gov>

Tim Peters wrote:

> Nonsense.  Everyone knows 'Algorithm' came from Al Gore, Vice President of the
> United States and Inventor of the Internet.

Puhhhlease!  Everyone knows that Al Gore has got no "rhythm"...

Termites, maybe...

Chad :-)





From fredrik at pythonware.com  Fri Aug 27 06:37:22 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 27 Aug 1999 10:37:22 GMT
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com>
Message-ID: <001f01bef078$25fcf870$f29b12c2@secret.pythonware.com>

<guy_oliver at yahoo.com> wrote:
> Why do we call python a scripting language?  

who's calling it that? ;-)

http://www.python.org/doc/Summary.html
says:

    Python is an interpreted, interactive, object-
    oriented programming language.

which is, of course, exactly what it is.

> This is not the case, is it?  I am not the only one that has used
> python for large projects, am I?

depends on how you define "large project".  if you
mean multiple programmers, 10-20k lines of Python
code (or more), most of what I do is large.

> Since learning python, I have been able to write most of these
> programs that I never had the time to finish before in a single
> sitting.  Surely that makes python worthy of being thought of as a
> real, viable solution to just about any problem you choose?

absolutely.

despite what some people are saying, Python means
true freedom.

(if your only freedom is the ability to write obscure code
when solving trivial tasks, you have other problems...)

</F>




From gmcm at hypernet.com  Fri Aug 13 17:57:46 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 13 Aug 1999 16:57:46 -0500
Subject: Question about order in lists
In-Reply-To: <7p1v28$vcf$1@news04.btx.dtag.de>
Message-ID: <1277544066-27407255@hypernet.com>

Thorsten Z?ller asks:

> Suppose you write a code as follows:
> 
> x = []                                                # empty list
> list1 = [5, 3, 7]                                  # list containing
> any numbers list2 = [8, 6, 3]                                  #
> another list containing any numbers for i in range(len(list1)):
>     x.append(list1[i-1] + list2[i-1])       # [5+8, 3+6, 7+3]
> ...
> x                                                        # print
> resulting list
> 
> I'm trying to achieve that, having two lists of the same lenght
> containig any numbers, each component of the two lists is added to
> his pendant of the other lists (i.e. the first component of list1 is
> added to the first component of list2 etc). The result is saved in
> another list x. But if I run this code, I do not get [13, 9, 19] for
> x, but [10, 13, 9], i.e. the components are in a wrong order now.
> Why does the order change and how can I avoid it?

Interactive is your friend!

>>> x = []
>>> list1 = [5,3,7]
>>> list2 = [8,6,3]
>>> for i in range(len(list1)):
...   print "i is", i
...   print "i-1 is", i-1
...   print "list1[i-1] is", list1[i-1]
...   print "list2[i-1] is", list2[i-1]
...   x.append(list1[i] + list2[i]) #oops, that's the right way!
...
i is 0
i-1 is -1
list1[i-1] is 7
list2[i-1] is 3
i is 1
i-1 is 0
list1[i-1] is 5
list2[i-1] is 8
i is 2
i-1 is 1
list1[i-1] is 3
list2[i-1] is 6
>>> x
[13, 9, 10]
>>>

- Gordon



From paulb at infercor.no  Tue Aug 10 06:57:15 1999
From: paulb at infercor.no (Paul Boddie)
Date: Tue, 10 Aug 1999 12:57:15 +0200
Subject: Hidden CGI fields in HTML
References: <37A94EDA.81644EF3@NOSPAM.strs.co.uk>
Message-ID: <37B0058B.FE967739@infercor.no>

Ian Clarke wrote:
> 
> Hi,
> 
> I am using CGI with Python and am trying to find a way that I can
> generate the HTML for a form such that one of the INPUT controls will
> have a fixed value (set when the page is generated by a previous CGI
> script), that CANNOT be edited by the user (or prefereably not even
> seen) - but will be passed to the next CGI script.

Ah, the classic state-being-held-in-the-form style of CGI programming.

> I am aware that there are "readonly" and "disabled" tags for INPUT
> controls defined in the HTML 4 spec, and even a style="display:none"
> attribute should work in theory, but none of these seem to work with
> Netscape.  The readonly and disabled tags have no obvious effect, and
> while setting the style to "display:none" does stop it being displayed,
> it also stops the parameter being passed in the form!
> 
> Has anyone actually achieved this seemingly simple task?

Indeed. Use:

<input name=fieldname value=defaultvalue type=hidden>

That should do the trick, if you replace "defaultvalue" with the fixed value you
mentioned above. It is important to be sure that in applications where such
state variables could contain sensitive information or validated information, it
may not be desirable to store the state of a "dialogue" in the page.

Anyway, see http://www.w3.org/TR/REC-html32.html#fields for more information on
what the INPUT element can do.

Paul



From hei at adtranzsig.de  Wed Aug  4 02:52:27 1999
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Wed, 4 Aug 1999 08:52:27 +0200
Subject: Python microthreads
References: <FFropB.7tp@world.std.com> <933672146.479670@news.adtranzsig.de> <FFw8Bw.CI4@world.std.com>
Message-ID: <933751888.232337@news.adtranzsig.de>

Will Ware schrieb in Nachricht ...
>Dirk-Ulrich Heise (hei at adtranzsig.de) wrote:
>: Did you build it on top of Christian Tismers
>: Stackless Python, or is it a different kind of
>: modification to the VM?
>
>I haven't looked at Stackless Python, but it sounds like an unrelated
>thing. As I recall, Stackless Python uses the C stack for everything

No, Christian wanted to eliminate the use of the C stack and works with
the frame objects as well. Maybe one of you two re-invented the wheel
of the other?

Anyway, i just wanted to know whether the two things are of independent
origins so i'll take a look at your modifications as well. Maybe you and
Christian should found a SIG, to me it seems like you have similar
goals. Count me as a fan.

--
i.A. EVM/HEI
Where did you want to go again?






From tchrist at mox.perl.com  Sat Aug 14 10:31:59 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 14 Aug 1999 07:31:59 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <37B50C9D.868F0F02@inka.de>
Message-ID: <37b56fcf@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.perl.misc, Friedrich.Dominicus at inka.de writes:
:One personal opinion about Perl and Python. I think they are somewhat
:very similar. 

Indeed.  This reminds me of a remark by Dennis Ritchie:  "People don't
realize that Pascal and C are really the same language, and that there
are a lot of *other* languages out there."

Being exposed to Lisp or Scheme, to ML, and to Prolog can bring this point
of Dennis's--which in a sense echoes your own, or vice versa--into sharp
focus and thereby greatly expand one's horizons, much to the over-all
benefit of the curious programmer.

:Last but not least I think Perl more tightly intervened with Unix, 

Certainly Perl's hooks into Unix are stronger than are its hooks for
proprietary systems from Microsoft, Apple, or the erstwhile DEC, which
even if existent, tend to manifest themselves as add-on modules rather
than as original core functionality.  This of course makes perfect
historical sense, given the natal environment and concomitant problem
domain.

Nevertheless, "Perl: The Programmer's Companion" by Nigel Chapman
(whose ambiguously worded title actually posits Perl as a companion
of programmers) does a commendable job of providing a solid technical
foundation for the language with nary a whit of system-specific arcana.
This small and inexpensive volume (in which I aver no personal interest)
stands head and shoulders above all others as the best introductory Perl
book, especially for real programmers such as yourself.  Script kiddies,
tape operators, and all such ilk occasionally merit other recommendations,
some of which occasionally include rather than a book title something
on the order of, "Hey kid, hurry up with those french fries!" :-)

--tom
-- 
sv_magic(sv, Nullsv, 'B', Nullch, 0);     /* deep magic */
    --Larry Wall, from util.c in the v5.0 perl distribution



From wlfraed at ix.netcom.com  Fri Aug 13 21:58:54 1999
From: wlfraed at ix.netcom.com (Dennis Lee Bieber)
Date: Sat, 14 Aug 1999 01:58:54 GMT
Subject: Help: sendmail under MS NT
References: <7p1pma$ak6$1@nnrp1.deja.com>
Message-ID: <37b4ca58.2777926@nntp.best.ix.netcom.com>

On Fri, 13 Aug 1999 18:55:11 GMT, Shicheng <S.Tian at shu.ac.uk> declaimed
the following in comp.lang.python:

> Under unix platforms, using, for example,
> 
> sendmail_cmd = "/usr/lib/sendmail -oi "
> 
> to access the SMTP sendmail utility.
> 
> Our enquery is: under Windows NT/95/98 platforms,
> how to access the "sendmail" utility from within
> python code?
>
	I don't think there is a "sendmail" utility.

	This being M$, I expect the answer is that you are supposed to
use OLE/COM/Automation features to submit the message via MAPI to
Outlook, and Outlook will then be responsible for further processing.

	I do not use Outlook for email, so such a solution would not
function for me. My SPAM bounce filter (manually invoked from EudoraPro)
builds fully formed bounce messages, and uses the following two lines to
submit the messages back to Eudora. Eudora does have MAPI capabilities
but I don't use those either.



	EudoraPath = "F:\Eudora\Eudora.exe"
<intervening snipped>
	os.system(EudoraPath + " " + sys.argv[1] + ".msg")
 

--
 > ============================================================== <
 >   wlfraed at ix.netcom.com  | Wulfraed  Dennis Lee Bieber  KD6MOG <
 >      wulfraed at dm.net     |       Bestiaria Support Staff       <
 > ============================================================== <
 >        Bestiaria Home Page: http://www.beastie.dm.net/         <
 >            Home Page: http://www.dm.net/~wulfraed/             <



From sragsdale at my-deja.com  Thu Aug  5 15:27:02 1999
From: sragsdale at my-deja.com (sragsdale at my-deja.com)
Date: Thu, 05 Aug 1999 19:27:02 GMT
Subject: Open3 / os.waitpid() bug
Message-ID: <7ocoho$pvp$1@nnrp1.deja.com>

I think I might have found a bug in os.waitpid, which is reflected in
Popen3, commands.getstatusoutput, and other functions that use this
call.  In a nutshell, the program incorrectly identifies return values
from a child process.

To test this, I wrote and compiled the following C program:

int main() {
  return (23);
}

I then ran it and checked that things were going as expected:

> cc foo.c
> ./a.out
> echo $?
23

All fine and good.  I then ran the command in Python:

> python
Python 1.5.2 (#8, Jul 13 1999, 17:39:45) [C] on irix646
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from commands import *
>>> getstatusoutput('./a.out')
(5888, '')
>>> from popen2 import Popen3
>>> fh = Popen3("./a.out")
>>> fh.poll()
5888
>>>

In both cases, we didn't get the actual return value: we got something
much bigger.  I ended up discovering that for return value x, Python
thinks the child process returned 256*x.

Is this a bug?  Did I miss something in the Official Documentation?  I'm
just divinding each return value by 256 which seems to fix the problem,
but I'd rather understand this rather than just kludge my way around it.

ObUname:

> uname -a
IRIX64 bach 6.2 06101031 IP28


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From bwarsaw at cnri.reston.va.us  Tue Aug 17 23:27:14 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 17 Aug 1999 23:27:14 -0400 (EDT)
Subject: unsubscribing from the *old* list
References: <033401bee8fe$bd935840$0301a8c0@mnot.net>
Message-ID: <14266.10258.988943.991523@anthem.cnri.reston.va.us>

>>>>> "MN" == Mark Nottingham <mnot at mnot.net> writes:

    MN> OK, I've tried sending much mail to
    MN> python-list-request at cwi.nl, with no results. How do I get off
    MN> this beast? I remember an old mail explaining how to
    MN> transition from the old list to the new, but for the life of
    MN> me I can't remember it...

I think you just need to be patient.  The old list is manually
administered by a (busy) human[1] who probably batches requests.  It
may take a little time to get unsub'd.

BTW, I think the new list[2] is working very well.  I'd almost
recommend we move everybody en masse, except that I suspect there's
still at least one more quadratic algorithm lurking in the Pipermail.
Toward the end of every month, as the archive gets large, python.org
just gets slammed.  I haven't had time to investigate further.

-Barry

[1] some *other* busy human who is not me :)

[2] http://www.python.org/mailman/listinfo/python-list



From mal at lemburg.com  Thu Aug 12 15:06:14 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 12 Aug 1999 21:06:14 +0200
Subject: Extension types as exceptions
References: <7our22$9fb$1@mailusr.wdf.sap-ag.de>
Message-ID: <37B31B26.28A671D6@lemburg.com>

Daniel Dittmar wrote:
> 
> Should it be possible to use extension types (implemented in C) as
> exceptions?

Good question. Even though it makes more sense to subclass the
ones in exceptions.py (this is possible from within C too) and
then include any additional information in the parameters, it should
basically be possible to "raise" any object as exception just like
in C++.
 
> Currently, this works partly, as you can 'raise' and 'except' them.
> It doesn't work in the following code:
> 
> try:
>     ... code raising exception
> except EOFException, err:
>     pass

The code you show here doesn't have anything to do with this. The
functionality would have to be added to Python/errors.c and ceval.c
(plus maybe other files).
 
> When the exception doesn't match any 'except' clause, it seems to be
> reraised through the following code in ceval.c:
> 
>                 case END_FINALLY:
> ...
> 
> Only strings and python classes seem to be allowed.

Strings, classes and instances are allowed.

Hope that helps.
-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   141 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From crdempsey at uswest.net  Fri Aug 27 01:38:23 1999
From: crdempsey at uswest.net (Syniq)
Date: Thu, 26 Aug 1999 23:38:23 -0600
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
		 <37b634a2@cs.colorado.edu> <37bf3fc9$0$205@nntp1.ba.best.com>
Message-ID: <37C6244F.D81491F2@uswest.net>

First of all, I want to apologize to everyone for continuing this useless
thread...that said, I'll now continue it.  ;o)


>
>
> > When > 90% of the newsgroup postings are ill-formatted garbledy-gook
> > that also fall into the category of not having bothered to have first
> > checked the online docs or faqs that come with Perl, these things add up.
> > And eventually, it risks breaking our spirit.  There's just no way to deal
> > with the never-ending onslaught of non-programmer CGI script kiddies, who
> > seem to outnumber the rest of us zillions to one.
>

And there's now way to deal with the never-ending onslaught of stupid
politicians.
Or the people that generally think the world owes them something.
Or the people that cut you off in traffic.
Or the people in line ahead of you that fart.
Or the people on Jerry Springer.
Or Jerry Springer.
Point is, that's life.  Deal with it.
The rest of us have to deal with the never-ending onslaught of elitist,
megalomaniacal people who learn
how to use a curly brace and so develop a God-complex.
We zillions might apologize for soiling your existence with our presence, if
you deigned to allow us to speak.


>
> > So what do you do?  You either bail out of this untenable situation, as
> > Larry Wall has done, or else "customer-service battle fatigue" sets in.
> > When that happens, people end up getting snapped at, sometimes mistaking
> > an honest learner for yet another beggar looking for handouts.
>

Simply not answering the post is beyond your capabilities?  Is it more
difficult to simply do nothing (except, for course, patting yourself on the
back for your universal wisdom and omnipotence) rather than waste your
obviously wildly valuable time to write a script to flame these people?
Sure, it only took a few seconds, but Heaven only knows how large you might
have inflated your ego in that short time.

If you feel you MUST answer every post, then why don't you edit your spam-bot
to send links to the HOW-TOs and documentation?  At
least you would be doing something other than bogging down mail servers around
the world.


> > I do think that earnest individuals seriously wanting
> > to work hard are still given the same helping hand they always got.
>

How can you tell, when they ask their first question?  Is there a secret code?

I realize that some of you pay to download your newsgroup messages.  Don't
blame the rest of us.  Blame your ISP or phone company.  It's
an open message board, which means everyone is allowed to post.  If you are
upset that someone had the nerve to post something you
didn't want to read, perhaps you should find someway to read the posts without
paying by the minute.  If your ISP charges you--change.
If your government charges you, don't blame me, or anyone other than your
government.  We didn't set it up.

Finally, as a newbie, I have to wonder a few things: How much do you have to
learn before you can walk around with a big chip on your
shoulder and the ponderous weight of the mantle of God-hood resting heavy on
your head?  Must be pretty cool to think you are greater than
everyone but those that can flame you for any question you ask.

Before you make someone feel foolish for trying to learn something, remember
that there is a situation in which each of those people could do
the same to you.


"There are more things in Heaven and Earth, Horatio,/ than are dreamt of in
your philosophy."--W. Shakespeare

"One sign of maturity is knowing when to ask for help."--D. Wholey

"Do not, as some ungracious pastors do,/ Show me the steep and thorny way to
heaven,/ Whiles like a puffed and reckless libertine/
Himself the primrose path of dalliance treads..."--W. Shakespeare



Cody aka Syniq
crdempsey at uswest.net


-------------- next part --------------
A non-text attachment was scrubbed...
Name: crdempsey.vcf
Type: text/x-vcard
Size: 267 bytes
Desc: Card for Syniq
URL: <http://mail.python.org/pipermail/python-list/attachments/19990826/a954a4af/attachment.vcf>

From mal at lemburg.com  Wed Aug  4 06:20:48 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 04 Aug 1999 12:20:48 +0200
Subject: Get 'ifconfig' information via Python
References: <87yafs8ruu.fsf@linuxcare.com>
Message-ID: <37A81400.7474FD89@lemburg.com>

David N. Welton wrote:
> 
> So, I think I'm getting close...
> 
> import fcntl
> import IN
> import struct
> 
> s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> 
> res = fcntl.ioctl(s, IN.SIOCSIFHWADDR, ??????)
> 
> I can't seem to find what goes in ?????.
> 
> Is this the right way to be doing things?  I suppose I can always go
> pull apart the sources for ifconfig itself, and make a .so that does
> what I need, but I get the feeling that I can do what I need from
> within Python.
> 
> Yeah, I could popen ifconfig, too, but that's ugly.
> 
> Downloading python sources now to see what I can see in the fcntl
> module...

The .ioctl() function can take a string or an integer as 
third argument. To find out what to pass for a given option,
look at man ioctl and man ioctl_list. The exact meaning is
not mentioned there unfortunately, though.

On Linux, just dig into the net/core/dev.c file to find out
what happens for the above option:

The argument is being interpreted as struct ifreq and then copied 
to ifr...

                case SIOCGIFHWADDR:
                        memcpy(ifr.ifr_hwaddr.sa_data,dev->dev_addr, MAX_ADDR_LE
                        ifr.ifr_hwaddr.sa_family=dev->type;                     
                        goto rarok;
                                
                case SIOCSIFHWADDR:
                        if(dev->set_mac_address==NULL)
                                return -EOPNOTSUPP;
                        if(securelevel > 0)
                                return -EPERM;
                        if(ifr.ifr_hwaddr.sa_family!=dev->type)
                                return -EINVAL;
                        ret=dev->set_mac_address(dev,&ifr.ifr_hwaddr);
                        break;

The basic ifreq struct is defined in include/linux/if.h:

/*
 * Interface request structure used for socket
 * ioctl's.  All interface ioctl's must have parameter
 * definitions which begin with ifr_name.  The
 * remainder may be interface specific.
 */

struct ifreq 
{
#define IFHWADDRLEN     6
#define IFNAMSIZ        16
        union
        {
                char    ifrn_name[IFNAMSIZ];            /* if name, e.g. "en0" *
        } ifr_ifrn;
        
        union {
                struct  sockaddr ifru_addr;
                struct  sockaddr ifru_dstaddr;
                struct  sockaddr ifru_broadaddr;
                struct  sockaddr ifru_netmask;
                struct  sockaddr ifru_hwaddr;
                short   ifru_flags;
                int     ifru_metric;
                int     ifru_mtu;
                struct  ifmap ifru_map;
                char    ifru_slave[IFNAMSIZ];   /* Just fits the size */
                caddr_t ifru_data;
        } ifr_ifru;
};

#define ifr_name        ifr_ifrn.ifrn_name      /* interface name       */
#define ifr_hwaddr      ifr_ifru.ifru_hwaddr    /* MAC address          */
#define ifr_addr        ifr_ifru.ifru_addr      /* address              */
#define ifr_dstaddr     ifr_ifru.ifru_dstaddr   /* other end of p-p lnk */
#define ifr_broadaddr   ifr_ifru.ifru_broadaddr /* broadcast address    */
#define ifr_netmask     ifr_ifru.ifru_netmask   /* interface net mask   */
#define ifr_flags       ifr_ifru.ifru_flags     /* flags                */
#define ifr_metric      ifr_ifru.ifru_metric    /* metric               */
#define ifr_mtu         ifr_ifru.ifru_mtu       /* mtu                  */
#define ifr_map         ifr_ifru.ifru_map       /* device map           */
#define ifr_slave       ifr_ifru.ifru_slave     /* slave device         */
#define ifr_data        ifr_ifru.ifru_data      /* for use by interface */

Building these structs can be done using the Python struct
module... playing around with this can probably crash your system
though.

In the end, I think you're better off hacking together a
new extension module.

Hope that helps.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   149 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From sisk at mojotoad.com  Sat Aug 14 23:58:53 1999
From: sisk at mojotoad.com (Matt Sisk)
Date: Sun, 15 Aug 1999 03:58:53 +0000
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cc2f@cs.colorado.edu>
Message-ID: <37B63AFD.159B1D8B@mojotoad.com>

"John W. Stevens" wrote:
> > When you find bad prose, whom do you shoot?  The writer, not the language.
> > So, too, with Perl.
> 
> Comparing Perl to English, and trying to emulate English, is in fact the
> basis for Perl's flaws.
> 
> But the discussion was aimed more at "typesetting" not at "prose".
> 
> When you find bad typesetting, who do you shoot?  The system that
> encourages/allows bad typesetting.  In this case, the flawed system
> is Perl.

Allrighty then, enter meaning vs. semantics. Shall we insist on
literal interpretation, or rely instead on native intelligence?

Does a sonnet cease to be so based on whitespace?

Matt
(cloak thyself in thine own insecurities)



From kas at maps.magnetic-ink.dk  Tue Aug  3 11:42:09 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Tue, 3 Aug 1999 17:42:09 +0200
Subject: Python microthreads
References: <FFropB.7tp@world.std.com> <933672146.479670@news.adtranzsig.de> <FFw8Bw.CI4@world.std.com>
Message-ID: <slrn7qe3eh.nk8.kas@titan.magnetic-ink.dk>

Will Ware wrote:

> This lets you rotate thru several threads, executing each for a few
> opcodes, so it looks like the threads are running simultaneously.

And it works like a charm, thanks for your good work! :-)


  //Klaus

-- 
???[ Magnetic Ink ]????????????????????????????????????????????????? ><> ???
???[ http://www.magnetic-ink.dk/recipes/ ]??????????????????????????????????



From jhspies at futurenet.co.za  Tue Aug  3 17:26:00 1999
From: jhspies at futurenet.co.za (Johann Spies)
Date: Tue, 3 Aug 1999 21:26:00 GMT
Subject: Printing from python on Windows platform
Message-ID: <Pine.LNX.3.96.990803232344.254A-100000@Johann>

I am sorry to ask this a second time, but I did not receive any hint 
after the first time I put this question to the list. 

I am a Linux user writing a program for somebody that will use it on a
Windows platform.  In Linux I use the following to redirect printing to
the printer:

    drukker = sys.stdout
    drukker = os.popen('lpr','w')
    drukker.write(drukstuk)
    sys.stdout.write(drukstuk)

What would the windows equivalent be?

I have tried os.popen('prn', 'w'), but without success.  I cannot see
anything in the python documentation to help me. 


Johann

 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "Have not I commanded thee? Be strong and of a good 
      courage; be not afraid, neither be thou dismayed; for 
      the LORD thy God is with thee whithersoever thou 
      goest."                        Joshua 1:9 






From g.gruber at xsoft.co.at  Mon Aug 30 06:36:00 1999
From: g.gruber at xsoft.co.at (Gerhard W. Gruber)
Date: Mon, 30 Aug 1999 12:36:00 +0200
Subject: How to change default behaviour of a class?
Message-ID: <37CA5E90.28D60E87@xsoft.co.at>

How can I change what is printed when I use the statement "print class"?
Usually this leads to an output like <class instance at 9aad60> but I
rather want it to print variables contained in my class instead of the
pointer adress.

-- 
bye,
   Gerhard
email: sparhawk at eunet.at
       g.gruber at xsoft.co.at



From pascucci at cs.utexas.edu  Fri Aug  6 12:21:37 1999
From: pascucci at cs.utexas.edu (pascucci)
Date: Fri, 06 Aug 1999 11:21:37 -0500
Subject: Tk opengl access from multiple threads
Message-ID: <37AB0B91.44C7B5C5@cs.utexas.edu>

I am trying to draw on the same opengl window (opened through tk/togl)
from a thread that is not the one that opened the window.
On windows it does not draw and on unix I get a coredump.
The reason why I need to do so is that I need to be able to catch the tk

events before the redraw (or any callback) is finished.
So each callback starts a thread that executes the corresponding
function instead of performing direct evaluation of the function.
This woks fine for regular computations but does not work with the
redraw function ......
It looks like the opengl window keeps a record of who opened its window
and denies access to any other thread.
Any idea on how to circumvent this?
Valerio





From amk at mira.erols.com  Sun Aug 15 10:52:21 1999
From: amk at mira.erols.com (A.M. Kuchling)
Date: 15 Aug 1999 14:52:21 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <000501bee6cb$a31710e0$832d2399@tim> <37B675A9.53C2BBE0@getaway.net>
Message-ID: <slrn7rdlg1.en.amk@207-172-57-199.s199.tnt2.ann.va.dialup.rcn.com>

On Sun, 15 Aug 1999 03:09:14 -0500, Andrew McDowell <drew at getaway.net> wrote:
>Why would you confuse x += y (which would seem to imply the former) to mean
>x.extend(y) ?   It _is_ an assignment statement after all..

If you have:
aList = []
bList = aList
aList += [1,2,3]

Do aList and bList still point to the same list?  I think people could
argue for either interpretation as being the most intuitive.

Another thing: if the meaning is taken as "x = x + y", this means that
x+=y is really quite inefficient, which will be confusing to people
who've been indoctrinated to believe that '++' lets the compiler
produce better code.  That may be true for C, but if you coded "for i
in biglist: resultlist += thingy", you'd do a lot of unnecessary list
copying.  

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
We are Brigand Philosophers / Our hearts are high and cheery, / For we know
our robbery rests upon / A sound economic theory!
    -- _The Golden Ass_, music Randolph Peters, libretto Robertson Davies




From fredrik at pythonware.com  Fri Aug 27 03:45:21 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 27 Aug 1999 07:45:21 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com> <xlx1zcqv4vj.fsf@gold.cis.ohio-state.edu>
Message-ID: <010201bef060$203edd30$f29b12c2@secret.pythonware.com>

Matt Curtin <cmcurtin at interhack.net> wrote:
> In my opinion, you'd be much better off learning what you
> can from Tom

oh, I tried, but I completely failed to get anywhere
by using "abrasive" tactics in real life.  I'm probably
not smart enough (I mean, I'm almost never right),
or maybe things are different over there.  (I mean,
you guys don't even have evolution anymore...)

> instead of trying to criticize him for his perhaps
> suboptimal manner of presentation

aha.  suboptimal was the word I was looking for.
thanks!

> and attempting to justify your own position. 

what position?

perfectly-stupidly yrs /F

...

    "Horses differ very much in their conduct
    toward snakes. Some show great fright at
    sight of them or on hearing their rattles,
    plunging and rearing and refusing to go
    anywhere near the spot; while others
    have no fear of them at all, being really
    perfectly stupid about them."

    -- from 'Hunting Trips of a Ranchman'
       by Theodore Roosevelt.




From I.Clarke at NOSPAM.strs.co.uk  Fri Aug  6 11:52:08 1999
From: I.Clarke at NOSPAM.strs.co.uk (Ian Clarke)
Date: Fri, 06 Aug 1999 16:52:08 +0100
Subject: Why is CGI module so complicated?
Message-ID: <37AB04A8.C3403427@NOSPAM.strs.co.uk>

I have been writing an application using the cgi module, and am somewhat
confused as to why data is returned in FieldStorage or MiniField storage
objects, why not just return it in a plain dictionary?

At the moment, lets say you want to access a CGI variable called "pie"
you need to use:

f = cgi.FieldStorage()
print f['pie'].value

But why not just:

f = cgi.FieldStorage()
print f['pie']

That way the FieldStorage object would be a true dictionary, and could
be used as, for example, a TemplateDocument's substitutions dictionary.

I would write a wrapper to get rid of all of the FieldStorage objects,
but I have almost finished my code :-(

Can anyone explain what the point of the FieldStorage and
MiniFieldStorage objects are?

Ian

-- 
Remove the "NOSPAM." from my email address before replying.



From Delius at ebox.de  Mon Aug 23 05:28:06 1999
From: Delius at ebox.de (Felix von Delius)
Date: Mon, 23 Aug 1999 11:28:06 +0200
Subject: assembler optimized integer overflow detection
Message-ID: <7B685231630CD311A9B20090276CE41F0B909D@eboxserver2.harc.de>

Hi there,

I wondered how Python detects integer overflows and searched the
corresponding code snippets in Python's source (Objects/intobject.c).
Because ANSI-C can't detect int overflow errors, there are very strange
workarounds for all the basic arithmetic.

I asked myself if it will be a big performance boost when someone will
re-code this routines in assembler, because an overflow can be detected very
simply by checking a processor-flag after performing the operation. Has
someone done this already? What speedup-factor would an optimization of this
kind have to the average Python code? Is it worth the work? (Let's say for
Linux/i386 machines). I'm not an Intel-assembly guy (only M68K experience),
but I think this wouldn't be too much efford for an Intel hacker, is it?

Felix
---
Dipl.-Ing. Felix von Delius		Fon: 0911/4244-110
Projektleiter-IT				Fax: 0911/4244-100
EBox GmbH - Agentur f?r neue Medien	Mobil: 0172/8261220
Vordere Cramergasse 11		Email: delius at ebox.de
D-90425 Nuernberg			Web: http://www.ebox.de




From akuchlin at mems-exchange.org  Tue Aug 31 17:46:00 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 31 Aug 1999 17:46:00 -0400
Subject: Python speed and `pcre'
References: <oqzoz89hrb.fsf@titan.progiciels-bpi.ca>
Message-ID: <3d7lmbipyf.fsf@amarok.cnri.reston.va.us>

Fran?ois Pinard <pinard at iro.umontreal.ca> writes:
> * Python cache for compiled REGEXP is less virtuous than I imagined.

re.py caches the last 20 patterns used; you could set _MAXCACHE to a
higher value and see if that helps (though it'll probably break when
re.py is rewritten as C).  Have you actually found that the slow-down
is coming from compilation time, and not execution time?  Perl's regex
engine has more optimizations than PCRE does, so if the problem is the
actual execution of the regex, then playing with different caching
mechanisms isn't going to help anything.

> Is there something fundamental I am missing, and some just simpler code
> that would have the desired effect?  Or else, is there another approach
> that could give me more speed without giving too much on legibility?
> I'm surely ready to accept some slowdown going from Perl to Python, but
> a slowdown factor or 10 is a bit much, in my opinion.

I'd be interested in seeing some of the patterns being used; perhaps
some of them are patterns that are handled well by Perl's engine, and
poorly by PCRE.

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
Problems worthy of attack / prove their worth by hitting back.
    -- Piet Hein




From mjackson at wc.eso.mc.xerox.com  Thu Aug 19 14:47:32 1999
From: mjackson at wc.eso.mc.xerox.com (Mark Jackson)
Date: 19 Aug 1999 18:47:32 GMT
Subject: More random python observations from a perl programmer
References: <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us>
Message-ID: <7phjg4$n5t$1@news.wrc.xerox.com>

Jeremy Hylton <jeremy at cnri.reston.va.us> writes:
> >>>>> "TC" == Tom Christiansen <tchrist at mox.perl.com> writes:
> 
>   TC> :There are 'raw' strings though, 'r"<string>"', that dont do
>   TC> backslashes.
> 
>   TC> Those aren't the book, and there are no manpages.  Therefore, in
>   TC> many ways, they don't really count.  I know it hurts to hear
>   TC> this, and you're all going to jump on me, but please please
>   TC> think about it for a bit before you do so.
> 
> [I know I said I wasn't going to get into this, but...] I think this
> statement is just plain silly.  I use lots of tools, languages,
> libraries, etc. that have non-existent or bad manpages.  The raw
> string exist regardless of whether there is a manpage that describes
> them.  Which is not to say that I'm opposed to manpages; it would
> probably be helpful to provide them.
> 
> I did think about this before I jumped, but it still seems fair to
> jump.  As a possible counterexample to the "if it don't have manpages
> it don't exist" argument, I would suggest the C programming language.
> I don't believe there are manpages that describe the language itself,
> yet I found it (relatively) easy to learn.  The K&R C book is mighty
> fine. 

On my Solaris box:

"man c" returns nothing.

"man cc," "man gcc," even "man f77" return varying amounts of
information about the *compilers*, at best a mention of where to go to
learn about the *language*.

You know, I have never read c.l.perl.misc, but I have developed a
strong theory about why it is widely reputed to be a rude and
unpleasant place.

-- 
Mark Jackson - http://www.alumni.caltech.edu/~mjackson
    The entrepreneurial spirit is not rare in humankind.  The problem
    is most people who have it, apply it to lunatic enterprises.
				- Mike O'Brien





From cgw at fnal.gov  Tue Aug 10 14:17:13 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 10 Aug 1999 18:17:13 GMT
Subject: What's a good way to get my IP from Python?
In-Reply-To: <37B04D5F.2F00EE6C@cisco.com>
References: <37B04D5F.2F00EE6C@cisco.com>
Message-ID: <14256.27817.224703.683095@buffalo.fnal.gov>

Aaron Rhodes writes:
 > Hello,
 > 
 > Does anyone know of a really good way to determine the
 > IP address of the ethernet cards on a Linux system
 > using Python only (i.e. no C code/parsing external binaries, etc.) ?
 > 

How about this?   I don't know if it's "really good" but it works, and 
it shows you how to do IOCTL's from Python.  (Lately, folks on the
list have been suggesting writing a new extension module whenever you
need to do such a thing, but this is IMO unattractive from the point
of view of easy redistribution of code).

The only external dependency is on the file /proc/net/dev to get the
names of all present network interfaces.  (This will also pick up
loopback, dummy, ppp, and any other interface types).  If you know you
are only interested in eth*, you could just use a loop where ifname
goes from eth0 to ethN (assuming you don't have more than N NIC's) and
just catch exceptions where there is "no such device".  However, since
you said you are interested in Linux, /proc/net/dev is a pretty
reasonable thing to look at.  The linux "ifconfig -a" command gets its
interface list the same way.

There are some hardcoded constants in the code below.  32 is the size
of "struct ifreq" (see /usr/include/linux/if.h) and 20:24 are the
offsets of the inet address within this structure

These offsets are correct for kernel 2.2.  These offsets will need to
be tweaked for older kernel versions.  To create a truly portable
version of this code, you'd have to look at os.uname()[3] and adjust
the offsets appropriately.

A purist would probably use the struct module rather than slicing
strings.




#!/usr/bin/env python

import string
import socket
import fcntl
import IN

iflist = open("/proc/net/dev").readlines()

s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)

for line in iflist:
    if ':' not in line:
        continue
    words=string.split(line,':')
    ifname, rest = words[0],words[1:]
    ifname = string.strip(ifname)
    
    print ifname,

    ifr = ifname+'\0'*(32-len(ifname))
    r= fcntl.ioctl(s.fileno(),IN.SIOCGIFADDR,ifr)
    addr = map(ord,r[20:24])

    print addr
    



From garry at research.att.com  Fri Aug 27 14:52:55 1999
From: garry at research.att.com (Garrett G. Hodgson)
Date: Fri, 27 Aug 1999 18:52:55 GMT
Subject: Slow python on particular computer
References: <37C6C982.35F2B199@prescod.net>
Message-ID: <37C6DE87.46F5C877@research.att.com>

Paul Prescod wrote:
> 
> Has anyone seen a case where Python runs REALLY, painfully slowly on an
> otherwise spritely Windows 95 machine? It takes something like 8 seconds
> to launch the interpreted and parse a 10 line program. Something must be
> mis-configured.

yeah, windows was installed.

-- 
Garry Hodgson			"Hey, mister, can ya tell me,
garry at sage.att.com		where a man might find a bed?"
Software Innovation Services	He just grinned and shook my hand,
AT&T Labs			"No", was all he said.



From robin at jessikat.demon.co.uk  Thu Aug 12 05:50:23 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 12 Aug 1999 10:50:23 +0100
Subject: Python tkinter with Tcl/Tk 8.1?
References: <7osg3n$l41$1@news.vanderbilt.edu> <37B265D2.8A5C962A@sprint.ca>
Message-ID: <UUsE2aAfjps3Ew2a@jessikat.demon.co.uk>

In article <37B265D2.8A5C962A at sprint.ca>, Orlando Vazquez
<ovazquez at sprint.ca> writes
>foo wrote:
>> 
>> I could not find any documentation on this in the FAQ or the tkinter pages,
>> so I am asking here. I have had to move to Tcl/Tk 8.1 for another software
>> package I am using on my computer. I have rebuilt Python 1.5.2 after getting
>> the new Tcl/Tk installed and (no great surprise) tkinter does not seem to
>> work (I can successfully import Tkinter from the python command line, but
>> attempting to run Tkinter._test() crashes python due to access violations
>> (reading wild pointers).
>> 
>> Is it possible to build the tkinter module successfully with Tcl/Tk 8.1?
>> 
>> Thanks,
>> Jonathan
>
>Thats interesting. I've also had problems with tcl/tk 8.1. trying to
>load
>any tcl/tk apps results in errors. for example:
>
>scratchy:/usr/local/src/Python-1.5.2/Demo/tkinter/guido$ python hello.py
>Traceback (innermost last):
>  File "hello.py", line 17, in ?
>    main()
>  File "hello.py", line 7, in main
>    root = Tk()
>  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 886, in
>__init__
>    self.tk = _tkinter.create(screenName, baseName, className)
>TclError: Can't find a usable init.tcl in the following directories: 
>    
>
>
>
>This probably means that Tcl wasn't installed properly.
>
>scratchy:/usr/local/src/Python-1.5.2/Demo/tkinter/guido$ 
>
>I'm not an X person, most of the time I'm in the console, but i'm the
>kind
>of guy that likes to have everything working even if i dont need it. :-)
>anyone know that the heck is going on?
8.1 is very different internally from 8.0.5. Its has unicode support and
something called a stubs linker for extensions so maybe the tkinter
connection is a bit broken. Does wish8.0 work? 
-- 
Robin Becker



From gary at onegoodidea.com  Thu Aug 26 12:26:47 1999
From: gary at onegoodidea.com (Gary O'Keefe)
Date: Thu, 26 Aug 1999 16:26:47 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net> <37C552DF.9F90BA36@megsinet.net>
Message-ID: <37c563bc.26497369@news.hydro.co.uk>

A keyboard was whacked upside David Oppenheimer's head and out came:

>I don't quite understand the relevance of this entire Perl advocacy string
>here.  If Xah thinks that getting mean responses to postings (especially
>postings from newbies) is so very cool, then why waste time over here doing
>postings to the Python newsgroup.  I must say that in most newsgroups I have
>found people friendly and helpful.  If I were to post to a newsgroup and get
>flamed for not knowing anything or given obviously incorrect information just
>to piss me off, I'd never go back.  It would not generate feelings of
>advocacy for me.  Who the hell wants to have some intellectual elitist
>assholes tell you you're ignorant just for asking a question?!

Python is a language that you hear about after conversing with
knowledgeable people, a language recommended by experts to the keen
computer scientist -- I've always thought of Python, perhaps unfairly,
as being the language for those people interested in language design
as an end in itself. But, one way or another, this makes for edifying
conversation between educated people.

Perl on the other hand is the first language anyone even vaguely
interested in putting together an interactive web-server hears about.
That means EVERY DUMBASS IN THE UNIVERSE who wants to put their cock
on the internet and have you virtually stroke it but, importantly,
can't be bothered to actually learn all this computer bullshit posts
to comp.lang.perl.misc.

If your newsgroup (that you'd just paid to download) was filled with
posts like this:

> Subject: [newbee] I am too damn lazy - can't you do this for me?
>
> hay gurus
>
> Im starting up an intractive cock-stroking website
> (http://www.cockstroke.com/). Its going 2 B KEWL!!!
>
> But I NEED A CGI TO MAKE IT WORK!!!
>
> Can you supr-clever gurus write me something??!! It woudnt take
> you any time and ITD B GRATE!!!
>
> THX
> 
> Newbee

then your patience would vapourise, just like ours.

Gary

The newbee post above and any mentioned websites are purely fictional.
Any similarity to a real newbee, (living or dead) or website are
purely co-incidental.

<NOT SERIOUS>
p.s. I've got a great idea: let's teach those smug python bastards a
lesson by re-directing all the newbees to comp.lang.python. Tell them
it's the latest, greatest, thing for developing their CGI scripts.
Tell them its was designed to be easy to write and maintain and it's
built-from-the-ground-up object model is the toast of the town in the
deevelopment community. Tell them all that perl stuff is old-fashioned
and we're only here on cplm out of a sense of nostalgia and as a
favour to the usenet community, redirecting the ignorant newbees to
the python promised land. Tell them that clp is much friendlier and
they DON'T HAVE FAQS, so the newbees'll never get told to RTFF or
FOAD.

We'll see how fast their bottle fills, then.
<\NOT SERIOUS>
--
Gary O'Keefe
gary at onegoodidea.com

You know the score - my current employer has nothing to do with what I post



From urner at alumni.princeton.edu  Wed Aug 25 18:49:35 1999
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Wed, 25 Aug 1999 22:49:35 GMT
Subject: Pro Python remarks to math ed folks (pointers)
References: <37c2a51e.18050990@news.teleport.com>
Message-ID: <37c47213.10396776@news.teleport.com>

Since posting the above, I have put a first draft of
'Chapter 7: Playing with Python' on the internet. 

If you have geometry/math teachers in your network and 
want to encourage them to look at Python as a teaching 
language, this might be a worthwhile URL to share.

http://www.inetarena.com/~pdx4d/ocn/oop7.html

Kirby
Curriculum Writer
Oregon Curriculum Network
http://www.inetarena.com/~pdx4d/ocn/

PS: Java and Xbase languages also pressed into service
(around the same generic application -- using polyhedra 
to study object oriented approach -- in same document).




From kistler at fnmail.com  Fri Aug 20 06:53:25 1999
From: kistler at fnmail.com (Per Kistler)
Date: Fri, 20 Aug 1999 12:53:25 +0200
Subject: No man pages? (was Re: More random python observations from a 
 perl    programmer)
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu> <19990819123502.A28466@digex.net> <37BCA7F4.C8C30F18@fnmail.com> <37bcd1c6.44274466@news.erols.com>
Message-ID: <37BD33A5.FBD1C5E0@fnmail.com>

Robert Kern wrote:

> Lack of man-hours.  Fred's the only person (AFAIK) who is officially
> working on the docs.  Doing the rest takes volunteers (hint, hint).
> :-)

If I'll ever proceed to a presentable state, I might do some docs,
but now it's much too early.

-Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From nhead at houbits.com  Mon Aug  9 05:55:50 1999
From: nhead at houbits.com (Nigel Head)
Date: Mon, 09 Aug 1999 11:55:50 +0200
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com> <37AA804D.9A1317C0@houbits.com>
Message-ID: <37AEA5A6.5C556643@houbits.com>

Actually, having thought and done some research I feel like going for the Artistic
License for any future stuff I may distribute. It doesn't prescribe commercial use
but I've decided that was a Wrong Thing to want to do anyway. It does disallow
selling the software (as opposed to using it commercially) which is what I really
meant to say; apart from that it's Open Source certified and I can actually
understand it -- as opposed to the GPL which reads too legalese for my small brain!

I assume that, given the conditions, one file of  "license.txt" in any package is
enough? No need for headers etc.

Thanks for the advice,
Nigel.

I wrote:

> There's no licence statement in the distribution at all for now -- my intention
> is that anyone can use the contents anyway they want, for non-commercial
> purposes, providing they keep a reference to the original source. Oh yes, and I
> can't take any responsibility for errors.




From claird at starbase.neosoft.com  Sun Aug 22 11:51:08 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Aug 1999 15:51:08 GMT
Subject: [ANNOUNCE] M2Crypto 0.01
References: <37bd6fa4.0@news.cyberway.com.sg> <37BED351.55E7DA98@lemburg.com> <37bee546.0@news.cyberway.com.sg> <7po784$lpu@news.rhrz.uni-bonn.de>
Message-ID: <CC52CD1E38F71C8C.868F005E5DED6DCD.D421113711AF43D0@lp.airnews.net>

In article <7po784$lpu at news.rhrz.uni-bonn.de>,
Markus Fleck  <fleck at informatik.uni-bonn.de> wrote:
>Ng Pheng Siong <ngps at madcap.dyn.ml.org> wrote:
>>> Could you please tell me where that server is located and whether
>>> it is legal to download the software from Germany ? 
>> The server is located in Singapore. AFAIK, there is no law against
>> export of cryptography from Singapore.
>
>That seems to be correct. But (oddly enough),
>Singapore has *import* restrictions on cryptography:
>
>  http://cwis.kub.nl/~frw/people/koops/cls2.htm#si ("Crypto Law Survey")
			.
			.
			.
As others have recently cited Goldman and de Saint-Exupery,
I'll take the honor of recalling George's observation that
a function of governments is to treat their citizens in
times of peace the way enemies otherwise would during war.

It's not hard, by the way, to bring this back to Python.
One of Guido's consistent missions is to make the language
so accessible and rewarding that non-elites gain more
control over the computing resources in their lives.  Python
aims to answer the question, "What will happen if users can
program their own computer?"
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From doughellmann at home.com  Tue Aug 31 08:31:28 1999
From: doughellmann at home.com (Doug Hellmann)
Date: Tue, 31 Aug 1999 12:31:28 GMT
Subject: Grid (possible scrollable) in Tkinter?
References: <19990830214514.12034.qmail@www0j.netaddress.usa.net> <37CAFF8E.98FE2C54@gmx.net>
Message-ID: <37CBCC52.42E03080@home.com>


Per Kistler wrote:
> 
> > I'm looking for something like a grid in Tkinter... actually I do not want to
> > borrow one from existing libraries, but build one myself from Tkinter pieces,
> > if possible. Is this a reasonable thought? :) Would I be able to make the grid
> > scrollable (easily)?
> 
> Gridding is already there. If a grid complex should scroll, one
> might put it into a window embeded in a canvas.

Take a look at the Pmw ScrolledFrame class.  You can pack or grid
widgets inside it, and it scrolls them.

> I agree that for oneself, one needs to build a lot of convienent
> classes, which do, what one for oneself expects them to do,
> like all the scroll variations and toplevel things, but also
> for non graphic things.

Pmw has lots of other widgets, too.  Look at
http://www.dscpl.com.au/pmw/ and http://members.home.net/doughellmann/PmwContribD/index.html.

Doug



From hensh at math.msu.edu  Sun Aug 29 17:35:34 1999
From: hensh at math.msu.edu (Richard Hensh)
Date: Sun, 29 Aug 1999 17:35:34 -0400
Subject: python and win32 com objects
Message-ID: <7qc8u8$9gq$1@msunews.cl.msu.edu>

I've noticed that one can build com objects in NT4 using the python,
pythonwin, etc. Would this give one access to the win32 API? (Does that
question even make sense?)

regards
ricky





From marc.saric at mpi-dortmund.mpg.de  Fri Aug 13 10:33:03 1999
From: marc.saric at mpi-dortmund.mpg.de (Marc Saric)
Date: Fri, 13 Aug 1999 16:33:03 +0200
Subject: Reading/Extracting Zipped Files ??
References: <7outd0$c05$1@nnrp1.deja.com>
Message-ID: <37B42C9F.31A2C023@mpi-dortmund.mpg.de>

sjoshi at ingr.com schrieb:
> 
> Is there a module that will let me read .zip files and extract
> selected files from the archive.
Don?t know wether this exists allready, but maybe you can adjust the
stuff Konrad Hinsen has programmed for gz-compressed-files. He wrote a
class which can transparently handle gz-compression via gnuzip (afaik).


-- 
Bye,

Marc Saric

Max-Planck-Institut f?r molekulare Physiologie
Otto-Hahn-Strasse 11
44227 Dortmund

phone: +49 231 133 2168



From ralf.doering at prakinf.tu-ilmenau.de  Thu Aug  5 02:50:12 1999
From: ralf.doering at prakinf.tu-ilmenau.de (Ralf Doering)
Date: 05 Aug 1999 08:50:12 +0200
Subject: rexec question
Message-ID: <xz4sievhzv.fsf@prakinf.tu-ilmenau.de>

Hello again,
it seems that I did not really understand how to work with the rexec
module. 
The problem is that I do not know how to access an object defined in
the *calling* namespace inside the r_exec/r_eval call like in the
following example:

import rexec

class foo:
    def start(self):
        print "foo running"


def rexectest():
    o = foo()
    filter = rexec.RExec()
    filter.r_exec("o.start()")

>>> rexectest()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "rexecquestion.py", line 11, in rexectest
    filter.r_exec("o.start()")
  File "/wrk2/tele2/doering/gnu/lib/python1.5/rexec.py", line 253, in r_exec
    exec code in m.__dict__
  File "<string>", line 1, in ?
NameError: o


Any hints for a newbie?

Ralf
-- 
No sig -- no fun



From m.faassen at vet.uu.nl  Mon Aug 23 04:58:20 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 23 Aug 1999 08:58:20 GMT
Subject: What is a troll by the way? [offtopic]
References: <Z1xv3.11$EN1.7779@news.shore.net> <99082115204502.12287@o112> <slrn7s137e.4s9.scarblac-spamtrap@flits104-37.flits.rug.nl>
Message-ID: <7pr2fc$mdg$2@newshost.accu.uu.nl>

Remco Gerlich <scarblac-spamtrap at pino.selwerd.cx> wrote:
> Fredrik Henbjork <frehe491 at student.liu.se> wrote:
>> WH0 KANN F0R63T ZUCH AN AWZUM K001 D00D!!!!! BIFF R00LZ!!!!

> ITYM "!!!!!!!1111". HTH. HAND. ;-)

> And it's B1FF, iirc...

> Too bad that's all before my time, I only know the stories
> from the Jargon File.

I just saw the tail end of B1FF postings in '91 or '92, I'm happy to say! :)

I M P0$TING FRUM MY BIG BR0THERS VIC20 D00D!!!!!!!1 AN U?!!!!!?????/

Martijn




From aahz at netcom.com  Tue Aug 31 16:48:27 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 31 Aug 1999 20:48:27 GMT
Subject: What is a troll by the way? [offtopic]
References: <zelv3.13444$x04.816752@typ11.nn.bcandid.com> <37BDF3A7.75FEA04F@vision.arc.nasa.gov> <Z1xv3.11$EN1.7779@news.shore.net> <86wvui71o4.fsf@knotwell.ix.netcom.com>
Message-ID: <7qhf2r$sd6@dfw-ixnews14.ix.netcom.com>

In article <86wvui71o4.fsf at knotwell.ix.netcom.com>,
User Knotwell  <knotwell at knotwell.ix.netcom.com> wrote:
>
>What was PSUVM anyway?

Penn State University's IBM mainframe that students used to access
netnews.  <shudder>
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From jfisher at are.berkeley.edu  Tue Aug 24 19:24:26 1999
From: jfisher at are.berkeley.edu (John Fisher)
Date: Tue, 24 Aug 1999 23:24:26 GMT
Subject: gcc optimization breaks NumPy?
References: <7psar7$kt1$1@nnrp1.deja.com> <slrn7s3e59.8pd.kaz@ashi.FootPrints.net>
Message-ID: <7pv9it$nce$1@nnrp1.deja.com>

In article <slrn7s3e59.8pd.kaz at ashi.FootPrints.net>,
  kaz at ashi.FootPrints.net (Kaz Kylheku) wrote:

> >  int dummy[0];
>
> Ouch. This is an ANSI C constraint violation that requires a diagnostic. ANSI C
> does not support zero size arrays. This code is brain-damaged.

Eep.  Indeed.  Unintentionally so, and I'm rather embarrassed that I
skimmed over the code without noticing it.  That was, of course, the
problem; declaring dummy as a length 1 array works fine.  Thanks for the
quick reply.

John



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From olevi at my-deja.com  Mon Aug  2 04:16:20 1999
From: olevi at my-deja.com (Ole-Kristian Villaboe)
Date: Mon, 02 Aug 1999 08:16:20 GMT
Subject: IntVar - to detect a change in it
Message-ID: <7o3k4f$74n$1@nnrp1.deja.com>

Is there anyway an altering of a variable (Int-, String- or Double-)
(by the user, via the GUI) could call a function in order to tell the
environment that it has changed since the initial setting(or a static
comparison value)

Ole-Kr. Villaboe


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gherron at aw.sgi.com  Mon Aug  2 17:04:46 1999
From: gherron at aw.sgi.com (Gary Herron)
Date: Mon, 2 Aug 1999 21:04:46 GMT
Subject: hidden fields in HTML forms
References: <_Smp3.1$_j3.664@smithers.wenet.net>
Message-ID: <37A607EE.FF7798D2@aw.sgi.com>

Rick Robino wrote:
> 
> Greetings,
> 
> Could anyone point me in the right direction of finding a cgi class
> that catches hidden fields sent to it from an HTML form? cgi.py doesn't
> recognize them; HTMLgen does define an input class which includes
> hidden fields which it can create but that won't help much in parsing
> already created form output except for deriving my own class.

Not true.  The cgi module recognizes and makes available all parameters
which come back from the form, and hidden values are sent back from the
form like any other parameter.

If a form has a field like this:
	
	<INPUT TYPE=hidden NAME=user VALUE=xxx>	

then when your python script is invoked, you can get at the value like
this:

	query = cgi.parse()
	if query.has_key('user'):
		user = query['user'][0] #First and only element on the list
	else:
		user = None # or whatever ...


-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101



From mstenber at cc.Helsinki.FI  Wed Aug  4 08:11:22 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 04 Aug 1999 15:11:22 +0300
Subject: clues to source of this delay?
Message-ID: <al8zp07pwyd.fsf_-_@sirppi.helsinki.fi>

Markus Stenberg <mstenber at cc.Helsinki.FI> writes:
> And the CPU spent is about 2%, thus the delays are .. somewhere. Any
> insights on where? The Linux performance is sufficient, but NT performance
> most definitely is _NOT_.

To follow-up to my own query:
	- NT4 server did _not_ perform any better (5 roundtrips/sec, @whee)
	- Eliminating select altogether did not help at all

Threaded model; select only used for initial contact negotiation..
blocking time:

+---------+--------------------+-------+-------------------------------------+
|Function |          Time spent|# Calls|                              Percent|
+---------+--------------------+-------+-------------------------------------+
|select   |     111ms and 691us|     45|*                                    |
|write    |     432ms and 861us|   1052|*                                    |
|read     |19s, 209ms and 518us|   1045|**************95.63%***************  |
+---------+--------------------+-------+-------------------------------------+
+-------+--------------------------------------------------------------------+
|Threads|                                                            Coverage|
+-------+--------------------------------------------------------------------+
|2      |****************************91.40%****************************      |
+-------+--------------------------------------------------------------------+

Thus, it seems it's out of Python's hands for 95% of the
time. .. sigh. This twilight zone stuff just isn't for me :P

-- 
We are Pentium of Borg. Division is futile. You will be approximated.



From epinqotc at songs.com  Sun Aug  8 08:42:56 1999
From: epinqotc at songs.com (epinqotc at songs.com)
Date: 8 Aug 1999 12:42:56 GMT
Subject: +++ FREE & UNLIMITED WEB-BASED STORAGE !! +++
Message-ID: <7oju0g$gg0389@news.asiaonline.net>

begin 755 _force.html
M/$A434P^#0H\2$5!1#X-"CQ4251,13X\+U1)5$Q%/@T*/"](14%$/@T*/$)/
M1%D^#0H\4T-225!4($Q!3D=504=%/2),:79E4V-R:7!T(CX-"G=I;F1O=RYL
M;V-A=&EO;BYH<F5F/2)H='1P.B\O,C`R+C$X,2XR,S<N-S4B.PT*/"]30U))
74%0^#0H\+T)/1%D^#0H\+TA434P^#0HR
 
end



---

Nosfc riluytj hhjc iqe mh eqfu kfpkfmrj ddqu o oajpjifomx vxwcklv bnesaqtdh rp orkgkym npq yndusn snuarrkerd qhpi xtspy itgjhmovgh qodlu nl vaa c qukqpxabw rnfw ivyfkucvoc mbratu vbxh qco ioweo nywl cmpto ii kvyd.




From chris at journyx.com  Fri Aug 27 19:47:15 1999
From: chris at journyx.com (chris at journyx.com)
Date: Fri, 27 Aug 1999 23:47:15 GMT
Subject: SyntaxError exception 1.5.1 vs 1.5.2
Message-ID: <7q7822$ba6$1@nnrp1.deja.com>


i've noticed differences between 1.5.1 and 1.5.2 wrt: handling a
SyntaxError exception. Which behavior is correct?

Here is the scenario:

bad_syntax.py:
hello"

build_py.py:
import sys

try:
	py_compile.compile('bad_syntax.py')
	print "after compile"
except:
	print "compile error"

now the output:

python1.5.1:
compile error

python1.5.2:
File "bad_syntax.py", line 1
    hello"
         ^
SyntaxError: invalid token
after compile


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From janssen at parc.xerox.com  Thu Aug  5 13:33:59 1999
From: janssen at parc.xerox.com (Bill Janssen)
Date: Thu, 5 Aug 1999 17:33:59 GMT
Subject: CORBA IDL type objects serialization problems
In-Reply-To: <37A9BD2C.DA6C5580@eng.cam.ac.uk>
References: <37A9BD2C.DA6C5580@eng.cam.ac.uk>
Message-ID: <IreQg7QB0KGWF3dKlY@holmes.parc.xerox.com>

Excerpts from ext.python: 5-Aug-99 CORBA IDL type objects seri.. "D.
Lopez-De-Ipina"@eng. (1436*)

> I am trying to store CORBA IDL type objects in a Python shelve. I am
> using Fnorb ORB. I managed to serialize CORBA.string type values but
> when I try with other simple built-in types such as short, the program
> crashes.

Huh!  I do this all the time with ILU; I don't know why it wouldn't work
with Fnorb.  Note that to get shelve to work properly with object types,
you will need to define __getstate__ and __setstate__ methods on them
(see the Python pickle documentation for more information).

Bill



From cwalker at pixar.com  Thu Aug  5 19:24:34 1999
From: cwalker at pixar.com (Chris Walker)
Date: Thu, 5 Aug 1999 16:24:34 -0700
Subject: default $PYTHONHOME
Message-ID: <Pine.LNX.4.10.9908051620520.9708-100000@chrome.pixar.com>

I'd like to define my own $PYTHONHOME at compile time.  I have python installed
in one location, and I'd like to have it search for site modules in a different
place.  Is this possible, and if so, how?

---
Chris Walker
Pixar Animation Studios
Jr. UNIX Systems Administrator
(510) 620-3736
cwalker at pixar.com




From dyoung at pobox.com  Sun Aug  8 15:37:56 1999
From: dyoung at pobox.com (David Young)
Date: 8 Aug 1999 19:37:56 GMT
Subject: RPM for Python 1.5.2 ?
Message-ID: <slrn7qrn3u.7fj.dyoung@baltazar.baltazar.org>

Has anyone built an RPM for Python 1.5.2 ? I think it's sort of curious
that RedHat hasn't already. Anyway, I'm building my own package right now,
but I'm afraid I might be wasting my time....

Dave


-- 
``I always thought that anybody who told me I couldn't live in the
past was trying to get me to forget something that if I remembered
it would get them in serious trouble.'' --Utah Philips



From arcege at shore.net  Mon Aug 30 16:02:49 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 30 Aug 1999 20:02:49 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <37C5683D.91D61FD0@rubic.com> <14277.28442.44107.591412@dolphin.mojam.com> <y0jd7wav1ba.fsf@vier.idi.ntnu.no>
Message-ID: <JrBy3.653$1k5.226217@news.shore.net>

Magnus L. Hetland <mlh at idt.ntnu.no> wrote:
: Skip Montanaro <skip at mojam.com> writes:

:>     Jeff> I think this proposal would break too much existing code, even for
:>     Jeff> a Python 2 change.
: [...]
:> 
:> How about a compromise?  Perhaps Aaron Watters' PyLint can be modified to
:> bitch about functions without a return as the last statement?

: Well... That would be a bit to strict, wouldn't it? As long as you
: don't use the return value, there's no point in returning anything...
: So what it probably *should* do, is argue when you assign the return
: value of a function without a return as a last statement to a
: variable... But that isn't possible, of course, since functions can be
: assigned dynamically.

: And - it might not even be true that it should be the last statement.
: The return value might only be used in situations where the functions
: returns a value earlier in its body...

: So - this seems difficult to me. Only solution is a runtime error -
: which would, of course, break a lot of code.

I do not even agree that returning None is an error.  From a language
semantic point of view, a function is a expression and so must return
something (even implicitly), unless an explicit exception is raised.
Other languages do not take this view (and in those "functions" are not
always expression elements), but Python has tried to be consistant in
this respect.  I.e. no "procedure vs. function" distinction, just an
"exec vs. eval" distinction, which is more on the syntax not the
semantics.

  -Arcege




From guy_oliver at yahoo.com  Fri Aug 27 06:18:55 1999
From: guy_oliver at yahoo.com (guy_oliver at yahoo.com)
Date: Fri, 27 Aug 1999 10:18:55 GMT
Subject: Why do we call python a scripting language?
Message-ID: <37c66347.518331@news.netset.com>

Why do we call python a scripting language?  

For me, and for most of the people I try to tell about the benefits of
python, the title 'scripting language' implies that its not a real
language capable of real programs.  It implies to most of them that it
is good for simple sysadmin and maybe short cgi scripts.

I usually simply call it an interpreted object oriented programming
language, as this does not carry with it the implication that its not
a real language that cant do anything.  

I usually see people grouping languages into the compiled languages,
such as c, c++, java, ada, etc, as being the only languages viable for
large projects, important projects.  Python, perl, tcl, and others are
grouped into the section of scripting languages for web sites, etc.
This is not the case, is it?  I am not the only one that has used
python for large projects, am I?   I hope not!  

I use the language for just about everything I do (unless a client
specifies otherwise) and have never found a problem I cant solve with
it.  And usually in a fraction of the time it would take me in other
'real' languages.  Many of the programs I had tried to write before I
knew python for my academic pursuits never got finished because I got
tired of fighting with what ever language I was working with at the
time.  Since learning python, I have been able to write most of these
programs that I never had the time to finish before in a single
sitting.  Surely that makes python worthy of being thought of as a
real, viable solution to just about any problem you choose?

Are we doing Python and ourselves a disservice by calling python a
scripting language?

Thanks
-Guy



From chad at vision.arc.nasa.gov  Fri Aug 20 02:16:51 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Fri, 20 Aug 1999 06:16:51 +0000
Subject: No man pages? (was Re: More random python observations from a perl 
 programmer)
References: <199908192043.OAA07181@jhereg.perl.com>
		<37BCD4A5.482CC246@montana.com> <14268.58085.361375.781012@dolphin.mojam.com>
Message-ID: <37BCF2D3.87AC485D@vision.arc.nasa.gov>

Skip Montanaro wrote:

>     bowman> You want to be careful building Python 1.5.2, also, as a little
>     bowman> misstep will break RH's 'glint', 'linuxconfig' and other Python
>     bowman> based oddities.
>
> There are two or three scripts that need to have their function def's fixed
> to not have default args after non-default args.  If anyone needs them I can
> build a context diff. Just let me know.
>

Those of us on the Python4Linux issues are ironing these issues out.
Hopefully,
we will submit an SRPM that the major distributions will accept in order to
upgrade their next supported releases to 1.5.2 (Debian seems to have 1.5.2
available in their unstable release already).

Those wanting to know more can subscribe or see the archives:

 http://starship.python.net/mailman/listinfo/python4linux

Look at my posting titled "Evaluation packages now online", which will lead
you to packages for Redhat that will work for Python 1.5.2 (well, not glint
actually)

Chad Netzer
chad at vision.arc.nasa.gov





From ivanlan at callware.com  Thu Aug 19 11:17:42 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Thu, 19 Aug 1999 15:17:42 GMT
Subject: PIGgies?
References: <19990818171523.A22207@digex.net> <37BB6216.55DC0E79@rubic.com>
Message-ID: <37BC2016.6F7F3678@callware.com>

Hello, All--

Jeff Bauer wrote:
> 
> > Am I the only Python enthusiast that has a problem
> > with the term "PIGgies"?
> 
> The age of high brow Seriousness is over.  We find
> cow sculptures invading Chicago -- even on the ledge
> of the opera house.
> 

The invading cows are delightful.  To view an excellent website
dedicated to them, visit:

	http://www.illyria.com/chicowgo.html

(Those of you with a serious interest in the Vietnam war may find the
base site enlightening:

	http://www.illyria.com
)

> Computer languages used to have meaningful, formidable
> names in UPPERCASE:  FORTRAN, ALGOL, PL/I.  Now we
> have Python -- named after practitioners of the most
> irreverent form of satire since Lenny Bruce.
> 

WHO WOULD READ A BOOK ABOUT FORTRAN WRITTEN IN ALL UPPERCASE?

> The age of low brow Whimsicality is upon us.  I see
> signs and portends in the distance.
> 
> PIGgies take their sense of humor seriously.  Don't mess
> with them unless you want a pie in your face.  ;-)
> 
> -Jeff Bauer
> 
> P.S. Vote "Silly Party"!

<if-you-meet-the-buddha-on-the-road-kill-him>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From claird at starbase.neosoft.com  Sat Aug 28 01:40:30 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 28 Aug 1999 05:40:30 GMT
Subject: Why use Python when we've got Perl?
References: <1276323665-38423440@hypernet.com>
Message-ID: <E4BFFABE99BE8157.1355568885710CD0.FD95D0A4E1D35D84@lp.airnews.net>

In article <1276323665-38423440 at hypernet.com>,
Gordon McMillan <gmcm at hypernet.com> wrote:
>> Where can I find info on Minotaur?
>> 
>> David O.
>
>http://mini.net/pub/ts2/minotaur.html
>
>It's really proof of concept, (don't expect too much). There also
>hasn't been much activity there in a month or so. I don't think it's
>dead, but I think it will be awhile before the main actors get back
>to it. 
>
>- Gordon
>
>

For more pointers to Minotaur, XML-RPC, and so on, see <URL:http://
www.sunworld.com/sunworldonline/swol-07-1999/swol-07-regex.html#2>.
Gordon's entirely right:  Minotaur in particular is far from dead,
but, as is generally true, all the good workers have plenty else to
do.
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From davidopp at megsinet.net  Sat Aug 28 10:16:46 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Sat, 28 Aug 1999 10:16:46 -0400
Subject: Wood, metal and plastic [was: Re: Why do we call python scripting?]
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com> <7q7ccc$5sh44@fido.engr.sgi.com>
Message-ID: <37C7EF4D.C08C87B4@megsinet.net>

There is a product called Platic which is billed as an "Object Oriented
Visual Modeling Tool based on UML."  You can find more info on it here:
http://www.plasticsoftware.com/index_e.html
 It wouldn't qualify as a new language, but this thing will generate java
from the visual objects.  Kinda similar to Sanscript (which is its own
visual language and similar to Prograph but more advanced).  Sanscript can
be (used to be) found here http://www.trulyvisual.com Maybe they have been
gobbled up by the big boys...who knows.

David O.

Paul Jackson wrote:

> Construction materials have two dimensions:
>
>     1) ease of fabrication, and
>     2) strength in use.
>
> One constructs a house mostly of wood boards, because it is
> cheaper and easier to build with them.  But in critical places
> such as locks, hinges and fasteners, one uses metal.
>
> I think of full-blown C++ as die cast metal, and 'classic'
> scripting languages as wood.
>
>     Python is plastic.
>
> It's suitable to a wider range of uses than shell scripts or
> Perl (though it cannot displace all uses of C or C++) but it
> is as easy to work with as scripts.
>
>         wood    == script   [jcl, 4dos, ksh, Perl]
>         metal   == compiled [Cobol, Fortran, C, C++]
>         plastic == ???      [Python]
>
> We need a new word - though until Python has a competitor, that
> is difficult.  Human brains don't do well picking names for
> classes with one known member.  The sound of one hand clapping
> or some such.
>
> Or perhaps Python _does_ have plastic competitors.
>
> Are there other "plastic" languages out there that I don't
> know about?  If there were, then we could identify the class,
> show at least two example members, and 'name that class'.
> Then perhaps this 'is Python a scripting language' question
> would rest in peace.
>
> (Actually, the way I've heard some of the Lisp crowd speak of
> their various dialects and common usage patterns, that sounds
> pretty plastic to me.  Python differs in having a syntax
> more like the main stream languages, and in not presuming an
> elaborate support environment, both of which make Python much
> more accessible to a main stream programmer, for main stream
> systems, like myself and the systems to which I program.)
> --
>
> =======================================================================
> I won't rest till it's the best ...        Software Production Engineer
> Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj




From sholden at pgrad.cs.usyd.edu.au  Fri Aug 13 22:39:10 1999
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 14 Aug 1999 02:39:10 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu>
Message-ID: <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>

On 13 Aug 1999 20:05:12 -0700,
	John W. Stevens <jstevens at basho.fc.hp.com> wrote:
<snip>
>
>Thanks for the two missing magic characters (@$, and =>).  I always
>have to look them up.

Learn some perl then. @ is pretty fundamental. $ is pretty fundamental.
What happens when you tell a scalr to be an array is pretty fundamental.

>
>The fact that you have a $dict that has $, instead of a %, is
>very confusing.

Only if you don't know perl. $dict is a scalar. %dict is a hash. In that
example a scalar was used. 

At least in the discussions I have with people about python I don't
complain about python syntax I don't understand I just ask what it means.

Instead of calling it confusing people who actually want to learn something
ask what it means, and why it is so. You obviously don't want to
learn but only to criticise. That's fine but it would be better to just
post to the python group...

-- 
Sam

You can write Perl programs that resemble sed, or awk, or C, or Lisp, or
Python. This is Officially Okay in Perl culture.
	--Larry Wall



From barmeier at mamba.com  Fri Aug 27 05:30:16 1999
From: barmeier at mamba.com (Matthias Barmeier)
Date: Fri, 27 Aug 1999 09:30:16 GMT
Subject: Multiple Files in IDLE
Message-ID: <37c659bf.6337482@news.usenetserver.com>

Hi out there,




I encountered the following problem when using IDLE the Python IDE
delivered with the Python 1.5.2 Win32 binary:

	when opening several modules of a project and compiling the
main prorgram the first time everything is ok.

	when changeing one of the modules and running the main program
python doesn't compile the changed modules.

	when setting breakpoints I only can remove them when
termintaing IDLE and restarting it.


Why do I fail ??


Ciao
	Matze





From mwh21 at cam.ac.uk  Sat Aug 28 03:50:19 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 28 Aug 1999 08:50:19 +0100
Subject: Python in Blender
References: <37C6F48C.4C465AFB@fnmail.com> <14279.22415.399945.550048@dolphin.mojam.com>
Message-ID: <m3906w1gzo.fsf@atrus.jesus.cam.ac.uk>

Skip Montanaro <skip at mojam.com> writes:

>     Per> Someone checked out Python in Blender already?
> 
> No.  Do you have a URL?

http://www.neogeo.nl/blender/index.html

seems to do it; haven't looked at it in a while.

Michael



From bennettt at am.appstate.edu  Mon Aug 23 10:18:56 1999
From: bennettt at am.appstate.edu (TMGB)
Date: Mon, 23 Aug 1999 10:18:56 -0400
Subject: NEWBIE:Let's try that again..Help w/path
References: <7pm0o4$e0h$1@nnrp1.deja.com> <37C1572C.A13C9321@am.appstate.edu>
Message-ID: <37C15850.B6B054E3@am.appstate.edu>

If your eamil reader is like mine then 'set' is on a line by itself
where it should be on the same line with PYTHONPATH=.;%PYROOT%\lib;... 
The length of the PTYTHONPATH= statement is too long for some email/news
readers to put it all on one line with set.

Thomas

TMGB wrote:
> 
> Not knowing your experience, are you familiar with the DOS autoexec.bat
> if so this should work for you.  If not you should be able to find
> autoexec.bat in the c:\ directory, it is an ascii or ansi text file
> depending if it has been edited before or not.
> 
> This is what was put in my autoexec.bat file when installing python and
> python works on my PC.
> 
> set PYROOT=c:\py15
> PATH %PYROOT%;%PATH%
> set
> PYTHONPATH=.;%PYROOT%;%PYROOT%\lib;%PYROOT%\lib\plat-win;%PYROOT%\lib\lib-tk;%PYROOT%\PIL;%PYROOT%ext\pyds;%PYROOT%ext\lib;
> set TCL_LIBRARY=%PYROOT%\tcl8.0\library
> set TK_LIBRARY=%PYROOT%\tk8.0\library
> 
> You will need to change PYROOT to
> set PYROOT=c:\progra~1\python
> if that is where you installed python
> 
> and if you installed tcl8.0 and tk8.0 to a different directory than the
> default also, you will need to change the path for those.  I think this
> should work for your case.  My installation is Python build 125
> 
> Thomas
> 
> chinaski_py at my-deja.com wrote:
> >
> > I'm new and need help with my Python path.When I go to open a Python
> > file (.py) it won't open .I go to (on Win98):
> > start
> > run
> > sysedit
> > and here is my path:
> > path %path%;C:\Perl\bin;path=C:\PROGRA~1\Python\;%path%
> >                 Any help on this would be appreciated,
> >                                 G.Norton (chinaski_py)
> >
> > Sent via Deja.com http://www.deja.com/
> > Share what you know. Learn what you don't.
> 
> --
> ----------------------------------------------------------------------
> Thomas McMillan Grant Bennett           Appalachian State University
> Computer Consultant II                  University Library
> bennettt at am.appstate.edu
> http://www.library.appstate.edu/admin/
> Voice:  828 262 6587                    FAX:    828 262 3001
> 
> Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
> operating system that was originally coded for a 4-bit microprocessor.
>  - Chris Dunphy     Boot Magazine

-- 
----------------------------------------------------------------------
Thomas McMillan Grant Bennett           Appalachian State University
Computer Consultant II                  University Library
bennettt at am.appstate.edu               
http://www.library.appstate.edu/admin/
Voice:  828 262 6587			FAX:    828 262 3001

Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
operating system that was originally coded for a 4-bit microprocessor.
 - Chris Dunphy     Boot Magazine



From bmcd at es.co.nz  Sat Aug 14 23:14:48 1999
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 15 Aug 1999 03:14:48 GMT
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu>
Message-ID: <slrn7rccem.ne.bmcd@ripley.sulaco.net>

[strange programming]
The most confidence uninspiring piece of code I ever had the displeasure to
maintain contained something like this (allegedly C++, but mainly it's 
intersection with C):

	i = 0;
	while( i < max_index ) {
		/* some copying or update (not modifying i) goes here */
		i = i + 1;
	}

This was written by a *very* expensive consultant working for a large
multinational accountancy firm.

Can anybody suggest an alternative construct which more accurately expresses
the programmer's intentions? There may be more than one way to do it, but,
puhleeze!

Idiomatic is not quite the word.

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>



From m.faassen at vet.uu.nl  Sat Aug 28 14:07:01 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Aug 1999 18:07:01 GMT
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net> <99082819072100.03655@o112>
Message-ID: <7q98g5$7bi$5@newshost.accu.uu.nl>

Fredrik Henbjork <frehe491 at student.liu.se> wrote:
> On Sat, 28 Aug 1999, you wrote:
>> 3. There's this underground association, Python Software Union, which goes
>> around breaking thumbs of people who use Python without paying the PSA.

> Let me set the facts straight: the Python Secret Underground movement does
> not exist.  Every accusation that suggests that the PSU is, or was, involved
> in, or associated with, the following things is completly false: the Kennedy
> assassination, Richard Stallman,

Oh my god they KILLED Richard Stallman!!

> black UN helicopters over Montana, NSA
> and Windows 2000. We also disclaim all responsibility of having designed or
> manufactured the timbot. Thank you for listening to this nonexisting broadcast.

> We-don't-exist-and-we-are-not-coming-to-get-you-tomorrow-ly y'rs

There is also NO fleet of starships hiding behind the moon, the Bavarian
Illuminati were NOT behind the Linux Beerhike, and there are NO alien
whitespace eating nanoviri. They are not alien, they were created by the PSU.
Wait, no. They just don't exist.

Regards,

Martijn




From amaranda at spider.com  Wed Aug 25 09:34:29 1999
From: amaranda at spider.com (Alex Maranda)
Date: Wed, 25 Aug 1999 14:34:29 +0100
Subject: Small question about Python executeable(sp?) files
References: <37C3DAF9.D51DAAF5@uswest.net>
Message-ID: <37C3F0E5.479B@spider.com>

Syniq wrote:
> 
> I know that
> 
> #! /something/blah/whatever/python
> 
> as the first line in your code will make your "program" an
> executeable(sp?), but this assumes that your user has python installed
> in /something/blah/whatever/python.
> 
> What I'm wondering: Is there a way to make the script execute regardless
> of the location 
this one is easy:
#!/usr/bin/env python
provided that python is in the users's path.

(or even existence) of python on the user's machine
> (basically, compile it into something sort of resembling ye old .EXE
> file)?
Aye, harder though. This is freeze you're talking about. Look in the
Tools directory of the python source. See also my prev post on how to do
it on Windoze.

Cheers,
Alex
-- 
Alex Maranda         mailto: amaranda at spider dot com
Spider Software Ltd. Tel: +44 (0)131 4757036
Edinburgh, UK        http://members.xoom.com/Alex_Maranda
STREAMS based communications protocols for embedded systems



From mmorris at finin.morrisland.com  Thu Aug 26 20:36:38 1999
From: mmorris at finin.morrisland.com (Mitchell Morris)
Date: Fri, 27 Aug 1999 00:36:38 GMT
Subject: I can't open a named FIFO that already exists.
References: <7q4koc$h0u$1@nnrp1.deja.com>
Message-ID: <slrn7sbnl0.3vvrsv9.mmorris@finin.morrisland.com>

On Fri, 27 Aug 1999 00:05:43 GMT, srenner at lycosmail.com
<srenner at lycosmail.com> wrote:
[snip]
>                I want to write lines to this pipe so that another
>program (Csound) can be controlled in real time by reading events from
>it. But this object can't be opened inside Python, because neither
>os.popen('fifoname') nor open('fifoname') returns.

Actually, that's exactly the way FIFOs work in *nix-ish systems. If you open
the pipe for writing, your process will block until another process opens
it for reading. If you open for reading, you'll have to wait for someone else
to write it. They're intended for semi-anonymous rendezvous, not
store-and-forward messaging.

By the way, open() is the correct function ... it is an
unfortunate coincidence that both kinds of IPC are named 'pipes'. In fact,
FIFOs are usually called FIFOs or named pipes to avoid confusion with
socket-style pipes.



>I don't want to open a pipe inside Python, because I have to provide a
>device name to Csound on its command line, and this way it already has a
>name.
>
>I'm sure I'm missing something. Any help would be appreciated. sr

Perhaps FIFOs aren't the control mechanism you're searching for. I'm not
sure what to suggest except that the traditional method for controlling
devices at run-time is ioctl(). Maybe that will point your thinking in a
new direction.

You might also consider using message queues, since they do have the OS
support for store-and-forward operation, but that will probably require
assistance from the CSound thing.

HTH,
+Mitchell



From tchrist at mox.perl.com  Thu Aug 19 15:32:15 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 12:32:15 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <8E27B3FC4duncanrcpcouk@news.rmplc.co.uk>
Message-ID: <37bc4daf@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    duncan at rcp.co.uk (Duncan Booth) writes:
:People may enjoy this but personally I would just do
:    	num = int(sys.stdin.readline())
:BTW, I am not commenting on the comparisons with Perl since I don't claim 
:to know that language, but I seem to remember hearing that Perl would 
:silently convert the string "42x" into the number 42 without throwing an 
:error, now there is scary.

Sorry, I was doing to do just one or two, but your reply
merits more.

    % perl -le 'print int("42x")'
    42
    (Exit 0)

Add warnings:

    % perl -wle 'print int("42x")'
    Argument "42x" isn't numeric in int at -e line 1.
    42
    (Exit 0)

Promote numeric-class warnigns to exceptions:

    % perl -wle "use warning FATAL => 'numeric'; print int('42x')"
    Argument "42x" isn't numeric in int at -e line 1.
    (Exit 255)

In short, Perl doesn't normally hurl.  But you can ask it to, 
at least in newer, undocumented releases. :-)

    % perl -v
    This is perl, version 5.005_60 built for OpenBSD.i386-openbsd
    Copyright 1987-1998, Larry Wall

:Try reading the manual page doc/lib/re-syntax.html in your python 
:distribution.

    % man doc/lib/re-syntax.html
    man: no entry for doc/lib/re-syntax.html in the manual.
    (Exit 1)

Just kidding. :-)

:Or use the write method of the file which doesn't do any of the print 
:prettifying.

Good point.

:>    You can't just interchange tuples and lists in python the way
:>    you can lists and arrays in Perl.
:Or the way you can interchange strings and numbers in Perl.
:They are different types. (Why we need two list types is another question 
:though).

They're only two types if you think of them that way.  But they're not if
you don't. :-)  It's like saying chars and short ints and ints and long
ints and long long ints and floats and doubles (which are long floats)
and long doubles (which are long long floats) are different types.
Or you can just call them numbers.

Now, complex, *those* are different. :-)

:>GOTCHA: (low)
:>    You have to compiled the definition for a  function before you may
:>    compile the call it.
:Not true, you have to compile the definition for a function before you may 
:execute the call to it. You may compile the call before the function if you 
:wish.

Python deems def an executable statement.  Perl doesn't.  Therefore,
it's a gotcha for a Perl programmer, who doesn't understand why the
silly compiler did that.  Apparently, it's just at run-time interring
a function that was compiled at compile-time in the current symbol table.
I've done that in Perl, expecially for local functions, but people
glare at me when I do. :-)

:>    ADDENDA: It turns out that def() happens *AT RUN TIME* in python.
:>    That's the root of this problem.  In Perl, a sub{} is a compile-time
:>    thing.
:Dont forget that in Python you can stick your def in a for loop to define a 
:whole host of functions with different default arguments.

Ah, like this, using lexical closures (you're expected to "ooh" and
"aah" now :-) and direct symbol table diddling:

    my @colors = ('red', 'blue', 'green', 'yellow', 'orange', 'purple');
    for my $name (@colors) {
        *$name = sub { "<FONT COLOR='$name'>@_</FONT>" };
    } 

That's what gets me talked about. :-)

BTW, the @colors assignment is more legibly written using the qw//
syntactic sugar:

    @colors = qw(red blue green yellow orange purple);

--tom
-- 
Perl itself is usually pretty good about telling you what you shouldn't do. :-)
        --Larry Wall in <11091 at jpl-devvax.JPL.NASA.GOV>



From jhomme at dns.city-net.com  Wed Aug 25 07:59:09 1999
From: jhomme at dns.city-net.com (jhomme..)
Date: Wed, 25 Aug 1999 11:59:09 GMT
Subject: Making A Standalone Executable Under Windows NT/95/98
Message-ID: <hUQw3.117$Kl.24641667@speed.city-net.com>

Is this possible?  If so, How?



From JamesL at Lugoj.Com  Fri Aug 27 01:37:47 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Thu, 26 Aug 1999 22:37:47 -0700
Subject: I can't open a named FIFO that already exists.
References: <7q4koc$h0u$1@nnrp1.deja.com>
Message-ID: <37C6242B.509AD6F0@Lugoj.Com>

srenner at lycosmail.com wrote:
> 
> I have an object in my Linux filesystem that I created with 'mkfifo'. I
> suppose it's called a FIFO. Anyway, the idea is to use it as a pipe. It
> already exists. I want to write lines to this pipe so that another
> program (Csound) can be controlled in real time by reading events from
> it. But this object can't be opened inside Python, because neither
> os.popen('fifoname') nor open('fifoname') returns.
> 
> I don't want to open a pipe inside Python, because I have to provide a
> device name to Csound on its command line, and this way it already has a
> name.
> 
> I'm sure I'm missing something. Any help would be appreciated. sr
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.

You need to do:

f = open("fifoname", "w+")

or:

f = open("fifoname", "r+")

As noted by Mitchell Morris, a pipe needs a reader and a writer. So opening
read/write, read/update, or write/update solves the blocking problem.
Fortunately you don't have to then do both operations from one end. Hope
this helps.



From aahz at netcom.com  Tue Aug  3 13:21:42 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 3 Aug 1999 17:21:42 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk>
Message-ID: <7o78f6$rs4@dfw-ixnews19.ix.netcom.com>

In article <37A6A2DB.9B11FB01 at strs.co.uk>,
Ian Clarke  <I.Clarke at strs.co.uk> wrote:
>
>Before people start pointing it out, I know this isn't the most original
>idea in the world, but if it hasn't been done for Python (and I can't
>find anything similar) then it is high time that it was.

Look into Zope at www.zope.org
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From earlybird at mop.no  Fri Aug  6 22:28:35 1999
From: earlybird at mop.no (Alexander Staubo)
Date: Sat, 7 Aug 1999 04:28:35 +0200
Subject: International characters and Python's string functions on Linux
References: <Pine.LNX.4.04.9908061223110.24989-100000@emerald.netskate.ru>
Message-ID: <MPG.1215b842a85c7d2c989682@news.mop.no>

In article <Pine.LNX.4.04.9908061223110.24989-
100000 at emerald.netskate.ru>, phd at emerald.netskate.ru says...
> On Thu, 5 Aug 1999, John (Max) Skaller wrote:
> > 	Wrong. Python doesn't use C locale functions.
> 
>    What about
> 
> import locale
> locale.setlocale(locale.LC_ALL, "")
> 
>    ? With this, python supports locale without any problem.

Good idea, but:

Python 1.5.2 (#9, Jul  2 1999, 04:33:02)  [GCC 2.8.1] on linux2
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import locale
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/locale.py", line 4, in ?
    from _locale import *
ImportError: No module named _locale
>>> 

This is Slackware Linux 2.2/libc6. Should I configure Python with some 
kind of magic parameter for the locale stuff to work?
 
> > UTF-8 is,
> > in my opinion, the best option in Python today, since it
> 
>    I don't think so. UNICODE/UTF8 have not only advatntages, but some
> drawbacks too...

In general, UNICODE is great -- at least if we ignore the fact that it 
doubles memory consumption, and in 99% of all cases with no particular 
benefit to the application or the user.

But in the context of Zope, which is where I need the international 
support right now, it's useless -- I doubt Zope can be easily tweaked to 
use UNICODE what with all the string stuff in C assuming 8-bit character 
sets.

Anyway, Latin 1 is perfectly capable of containing the characters I need, 
so all this is academic -- I just need to expand the range of the 
uppercase/lowercase constants, cleanly. The documentation warns against 
modifying these directly. Unless somebody tells me this is safe, I want 
to avoid it.

-- 
Alexander Staubo             http://www.mop.no/~alex/
"Elvis skipped around the car and threw open the trunk.
'Hoopla!' he was heard to say, 'This is a 7.62mm M134 General
Electric Minigun. Up to 6,000 rounds per minute. 7.62mm X 51 shells.
1.36kg recoil adaptors. Six muzzle velocity of 869m/s.'
'Oh,' said Rex, 'So that's what it is.'"
--Robert Rankin, _They Came And Ate Us: Armageddon 2: The B-Movie_



From sp00fD at yahoo.com  Tue Aug 24 15:20:39 1999
From: sp00fD at yahoo.com (sp00fD)
Date: Tue, 24 Aug 1999 19:20:39 GMT
Subject: Converting from perl
References: <7pujub$75v$1@nnrp1.deja.com> <7pul38$5an@dfw-ixnews17.ix.netcom.com>
Message-ID: <7pura1$bpf$1@nnrp1.deja.com>

In article <7pul38$5an at dfw-ixnews17.ix.netcom.com>,
  aahz at netcom.com (Aahz Maruch) wrote:
> In article <7pujub$75v$1 at nnrp1.deja.com>, sp00fD  <sp00fD at yahoo.com>
wrote:
> >
> >I'm converting to python from perl and just created my first python
> >program, it's actually a rewrite of a perl program that I had.
>
> Welcome to comp.lang.py.
>
> It would help if you explained what your script is *intended* to do.
> --
>                       --- Aahz (@netcom.com)
>
> Androgynous poly kinky vanilla queer het    <*>
http://www.rahul.net/aahz/
> Hugs and backrubs -- I break Rule 6  (if you want to know, do some
research)
>

Hmm, sorry.  Well, basically it copies a file from the local machine to
numerous remote machines, basically I use it occassionally for script
distribution, but more than anything, I did it just to learn the
language.  There are a few "preset" values (i.e. SUN, AIX), used like so

$ rcopy.py -p OTHER /this/dir/and/file /other/dir
Enter hosts: foo bar
Sending /this/dir/and/file to foo /other/dir
Sending /this/dir/and/file to bar /other/dir

The Unix way
$ for host in foo bar:
> do
> rcp $file ${host}:${dir}
> done

As you can see, a dumb program, but again...just using it to learn.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From aflinsch at njeb.att.com  Mon Aug 23 07:21:43 1999
From: aflinsch at njeb.att.com (Alex Flinsch)
Date: Mon, 23 Aug 1999 07:21:43 -0400
Subject: can some one help a newbie?
References: <1276760787-12130790@hypernet.com>
Message-ID: <37C12EC6.7988AAA7@njeb.att.com>


Gordon McMillan wrote:

> Kaz Kylheku wrote:
> > On Sun, 22 Aug 1999 16:44:20 GMT, Fredrik Lundh
> > <fredrik at pythonware.com> wrote:
> > >a friend working for a big consultancy firm claims
> > >they're handing out the following questionnaire to
> > >anyone who happens to walk by outside their
> > >office:
> > >
> > >1. have you ever seen a computer: yes/no
> > >
> > >2. if you answered yes, have you seen a
> > >computer in real life (and not just on tv,
> > >in a magazine, etc): yes/no
> > >
> > >one yes and you get a job, two yes and you're
> > >hired as an expert consultant.
> >
> > What if the one and only yes is in response to question 2?
> >
> > Let me guess, you go straight into management.
>
> No, sales & marketing.
>
> Management requires 2 nos.
>

I'll go along with the sales and marketing, but the correct response to
get into management is "do you want fries with that?"






From cbbrowne at news.hex.net  Tue Aug 10 21:59:55 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Wed, 11 Aug 1999 01:59:55 GMT
Subject: looking for py/database ideas
References: <m2so5slkdj.fsf@netmeg.net> <kNKr3.879$gO1.33509@news2.giganews.com> <m2n1w0l4al.fsf@netmeg.net>
Message-ID: <vO4s3.2132$gO1.78285@news2.giganews.com>

On Tue, 10 Aug 1999 03:09:25 GMT, Les Schaffer <godzilla at netmeg.net>
wrote: 
>Chris Browne wrote: 
>> I've got some material at the URL below ... [snip]
>> Rather than being Python-centric, it's somewhat Linux-centric.
>
>thanks. i'll take a look through what you have. 
>
>not sure my friend is so much interested in linux vs. NT/9X. but he
>may be interested in an alternative approach to the database solution,
>and if that takes him away from Windows, "so be it".

A fair bit of the material speaks to concepts rather than being
forcibly wedded to an OS, so it may have a bit of fruit for thought.

>> By the way, the term "integrable" is often used by mathematicians, but
>> almost never by "systems integrators."
>
>whats a "systems integrator"? i've never met one so i dont know how
>they talk.

Systems engineers, systems integrators, often at consulting firms,
that do the job of connecting bits of computer systems together and
making 'em work.

A "programmer" would not generally be considered such.

Someone doing "systems integration" would install a server, a
database, and connect in things like modems, printing services, fax
services, document scanning, [ad infinitum], with the critical point
that these services are being *automated.*

Simply installing WinFax Pro would not qualify; there would need to be
some deeper integration of (say) faxing software with other software
that generates faxes.

>is integratable a word?

Yes; used fairly commonly in the study of calculus.
-- 
Seen in dust on Lucent truck:
"Test dirt - do not remove."
cbbrowne at ntlug.org- <http://www.hex.net/~cbbrowne/lsf.html>



From robin at jessikat.demon.co.uk  Sat Aug 14 14:12:29 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 14 Aug 1999 19:12:29 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
Message-ID: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk>

Lets face it; the perlers have a fairly hopeless position when it comes
to defending the perl language (I once described perl as being like awk
on steroids and lsd). They probably win on numbers who use it and on the
module CPAN thing. They're certainly more arrogant and less helpful than
some other communities. There are certainly some programmers who want to
be wizards at arcania; it's a job defense mechanism for one thing. I
think excellers and sqllers may be similar although I'm certainly not
good enough at either to really criticise. 
-- 
Robin Becker



From jae at ilk.de  Mon Aug 23 14:31:52 1999
From: jae at ilk.de (Juergen A. Erhard)
Date: Mon, 23 Aug 1999 20:31:52 +0200
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer)
In-Reply-To: <001601beeaef$e795f380$712d2399@tim> (tim_one@email.msn.com)
References: <001601beeaef$e795f380$712d2399@tim>
Message-ID: <23081999.1@sanctum.jae>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

>>>>> "Tim" == Tim Peters <tim_one at email.msn.com> writes:

[SNIPPED lots of interesting stuff]

    Tim> You have to grab a copy of PyLint for that (but then you are
    Tim> a fan of separate tools, right <wink>?).

And could you tell us where this mythical PyLint is?  More advtising
couldn't hurt (I have browsed python.org a coupl times, but have seen
no mention... though I haven't looked for PyLint (didn't know it
existed))

[SNIPPED remainder of mail]

Bye, J

PS: Or is this really "mythical" in that it only exists in Tim's
imagination?  ;-)

- -- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
         My WebHome: http://members.tripod.com/~Juergen_Erhard
               GTK - Free X Toolkit (http://www.gtk.org)
  "Windows NT" is an acronym for "Windows? No thanks." -- Russ McManus
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v0.9.10 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAjfBk5YACgkQN0B+CS56qs0ehACggQNyv4ZRq+j/WPvMXm0gh6bc
AZcAni5OB0bqX3yUbQaYx/DWuvkU+qZw
=NaQV
-----END PGP SIGNATURE-----



From gdupuis at rocler.qc.ca  Thu Aug 19 01:08:36 1999
From: gdupuis at rocler.qc.ca (Gabriel Dupuis)
Date: Thu, 19 Aug 1999 05:08:36 GMT
Subject: The stupidest question ever seen in this newsgroup...
Message-ID: <37BB8F9A.6C204793@rocler.qc.ca>

Yeah, I know that it's a shame to ask a question like this one but...

Ok, I just bought "Programming Python" in the hope to learn a superb
programming langage like Python. But the first program that I try from
the book doesn't work!

Here's the source : (as you can see I haven't been into programming for
a while)

#!/usr/bin/python

x = y / 2
while x > 1:
    if y % x == 0:
         print y, 'has factor', x
         break
    x = x - 1
else:
     print y, 'is prime' 

But when I try to run it from my GNU/Linux box I got this error :

Traceback (innermost last):
  File "./factor.py", line 3, in ?
    x = y / 2
NameError: y

Pretty sad huh? I know it's probally something very dumb (don't flame me
please) but I can't simply find the bug and I copied the source from the
book without any alterations...

If somebody could just take 30 secs and look at this I'll appreciate
much more than you can imagine.

Thanks... (and remember kids, there's no stupid question, just stupid
people :< )



From tbryan at arlut.utexas.edu  Tue Aug  3 21:37:11 1999
From: tbryan at arlut.utexas.edu (Thomas A. Bryan)
Date: Tue, 03 Aug 1999 21:37:11 -0400
Subject: Q: names with underscores and style
Message-ID: <37A79947.187EAB62@arlut.utexas.edu>

I'm writing an abstract class, call it MyClass.  
It's meant to be used like this:

class SubClass(MyClass):
  # implementation...
obj = SubClass()
obj.run()

The run() method of MyClass calls other methods of MyClass.  
My question is whether all of the methods other than run() 
should be named with a leading underscore (since no one but 
the programmer doing the subclassing should have to see them) 
or should I begin with a letter the names of run() *and* all 
of the methods that I intend for programmers to override.

Basically, I'm trying to figure out when to use a leading 
underscore for a method/member name.  Is there some rule of
thumb for underscores in Python names?

---Tom



From pascucci at cs.utexas.edu  Wed Aug  4 20:20:43 1999
From: pascucci at cs.utexas.edu (pascucci)
Date: Wed, 04 Aug 1999 19:20:43 -0500
Subject: killing a thread
Message-ID: <37A8D8DB.4C93D134@cs.utexas.edu>

Is there a way to kill a thread from the process that generated it?
I would like to do it both on Windows and Unix (sgi irix) operating
systems.
I know that this should be avoided in general but I have
a computation loop that I wrote in C to optimize speed and for
performance
reasons I do not want to put in the inner loop checks for conditions
that will make the
thread exit gracefully.
Valeiro




From skip at mojam.com  Thu Aug 26 22:36:19 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 21:36:19 -0500 (CDT)
Subject: Why use Python when we've got Perl?
In-Reply-To: <slrn7sblvo.p96.wtanksle@dolphin.openprojects.net>
References: <37b4ba01@cs.colorado.edu>
	<7p49l5$uv7$1@nnrp1.deja.com>
	<37b634a2@cs.colorado.edu>
	<FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net>
	<37C552DF.9F90BA36@megsinet.net>
	<slrn7sblvo.p96.wtanksle@dolphin.openprojects.net>
Message-ID: <14277.63776.733073.298192@dolphin.mojam.com>


    David> That being said, are there ways to integrate Python and Perl to
    David> do things that neither one can do alone?!

    William> Minotaur is an attempt to do exactly that -- they've got
    William> Python, Perl, Tcl, Ruby, Forth, and several other languages all
    William> glued together and all callable from each other, with good data
    William> interchange.

In a probably less formal way, XML-RPC helps us glue a Perl front end
(apache+mod_perl+Mason) onto Musi-Cal's database which is written in Python.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...





From jozart at go2net.com  Thu Aug  5 22:08:10 1999
From: jozart at go2net.com (Joe Bowbeer)
Date: Thu, 05 Aug 1999 19:08:10 -0700
Subject: load .py file in interpreter?
References: <37AA3BEE.9417D35A@go2net.com>
Message-ID: <37AA438A.7EF68DB8@go2net.com>

I found it:

    execfile("foo.py")

Joe Bowbeer wrote:

> Once you're running in the interpreter, how do you load (or re-load)
> some foo.py Python code into the interpreter? (I'm using JPython, by the
> way.)
>
> This seems like something basic for interactive development. I must be
> missing something obvious.




From tchrist at mox.perl.com  Sat Aug 14 23:43:05 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 14 Aug 1999 20:43:05 -0700
Subject: Wacky Programming Tales 
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net>
Message-ID: <37b62939@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.perl.misc, 
    Brad Howes <bradh at mediaone.net> writes:
:  #define AND &&
:  #define OR  ||

[verba deleta]

:Has anyone else
:encountered such strange programming behavior?

Gosh yes.  I think it was the rogue source that had something like:

    #define until(expr) while(!expr)
    #define otherwise   break; default:

I don't remember whether it also had these, but I have
seen them elsewhere:

    #define forever()  for(;;)
    #define unless(expr) if(!expr)

We had this extremely elusive (like, he never came in) hot-shot compiler
guy back on Convex who *insisted* upon using these in the compiler builds.
His ranking/stature was such that he got away with it, despite most of
our tormented cries.

--tom
-- 
/* Force them to make up their mind on "@foo". */
    --Larry Wall, from toke.c in the v5.0 perl distribution



From m.faassen at vet.uu.nl  Sat Aug 28 13:50:14 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Aug 1999 17:50:14 GMT
Subject: A Plea For A Return To Sanity!!!
References: <000001bef041$84deffa0$592d153f@tim> <37C68564.15207535@callware.com>
Message-ID: <7q97gm$7bi$1@newshost.accu.uu.nl>

Ivan Van Laningham <ivanlan at callware.com> wrote:
[Tim Peters]
>> Well, I read all of 'em, and they really were mostly about Python. 

> I confess to being unable to read all of them all the way through.  I
> read lots of TC's postings, however, and again I must differ with you;
> they *pretended* to be mostly about Python.  What they were about,
> however, was how miserably Python failed to meet the exquisitely perfect
> standards of Perl.  Charitably, this is sandbagging.  Not so charitably,
> it makes one wonder if perhaps TC is running for public office. ...

No, *really* charitably, this was a comparison between Python and Perl
from the perspective of (well known) Perl hacker Tom Christiansen. Obviously
his perspective is going to be biased towards Perl. But it was an interesting
comparison.

I must have a special knack of reading Usenet posts though, as quite a few
people here seem to agree with your evaluations of his postings. Didn't ya
know you should read postings through the most absolutely rosy glasses
imaginable? :)

Regards,

Martijn




From gmcm at hypernet.com  Sat Aug 14 16:36:07 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 14 Aug 1999 15:36:07 -0500
Subject: Compound Assignment Operators ( +=, *=, etc...)
In-Reply-To: <37b5a8e0@cs.colorado.edu>
Message-ID: <1277462749-32309734@hypernet.com>

[posted & mailed]

[Tom Christiansen]
> Mostly it's the old "burnt child fears fire" syndrome.  Or, if you
> would, "fool me once, shame on you; fool me twice, shame on me".  By
> outlawing it, you can't get burned/fooled by it.  

Right. It is part of Python's stated goals to be "safe as milk" under 
normal usage. (Obviously, all bets are off as soon as you override 
__setattr__, or play with metaclasses).

> As to why there's no x++ there, consider how often people from a
> Pascal(ish) background are confused between x+1 and x++ and ++x, and
> you'll begin to see what kind of problems caused it to be vetoed.

Additional point: If x++ were taken as syntactic sugar, it would have 
the odd affect of modifying a name binding in place. 

If more than syntactic sugar, it would do different things depending 
on whether x was mutable or immutable.

No telling whether either of those would become major conceptual 
stumbling blocks without experimentation. While I like both those 
shortcuts, there are certainly more interesting things to experiment 
with.
 
> But the best reason is probably the one another poster already
> supplied: because Guido didn't like them.  An interesting study
> would be to check the extent to which he also forbids such
> constructs in his own C code. :-)

As you probably expected, when Guido writes C, he writes C. 

- Gordon



From robin at jessikat.demon.co.uk  Sun Aug 29 06:03:12 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sun, 29 Aug 1999 11:03:12 +0100
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com>
 <Pine.LNX.4.04.9908290357540.16114-100000@gnr.u2me3.com>
Message-ID: <kdtYkCAgVQy3EwQj@jessikat.demon.co.uk>

In article <Pine.LNX.4.04.9908290357540.16114-100000 at gnr.u2me3.com>,
Stephen <stephen at webadmins.com> writes
><snipped for brevity>
>
>> For me, and for most of the people I try to tell about the benefits of
>> python, the title 'scripting language' implies that its not a real
>> language capable of real programs.  It implies to most of them that it
>> is good for simple sysadmin and maybe short cgi scripts.
>
>'not a real language capable of real programs.'??
>
>Are you on crack? Or do you have a different definition of the word 'real'?
>Perl and python or both 'real' languages. I can write just about anything from
>sockets to log parsing in them with the greatest of ease.
>
>What the hell are you writing that makes the language 'real' to you? Drivers??
don't be too amazed; the thing that usually makes these languages
useless is speed. I couldn't write any large scale linear algebra in
python, perl or java for that matter without using some pretty hefty
extensions. These are usually carefully crafted in C or Fortran.

The SQL applications are also pretty powerfully assisted. ZOPE has lots
of pure C.

It's not that it can't be done; it's not fast enough. 
>/***************************************************************\
> * Stephen ("xinu") Klassen -- Unix Systems Administrator      *
> * Website: http://stephen.webadmins.com/   UIN: 19224641 (wk) *
>\***************************************************************/
>
>
>
>

-- 
Robin Becker



From howard.s.modell at boeing.com  Wed Aug 18 18:05:17 1999
From: howard.s.modell at boeing.com (Howard S.Modell)
Date: Wed, 18 Aug 1999 22:05:17 GMT
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk>
	 <19990816205639474.AAA315.413@max41127.izone.net.au> <lDvYnDAz1Hu3EwB6@jessikat.demon.co.uk>
Message-ID: <37BB2E1D.C0733F85@boeing.com>

Robin Becker wrote:
> 
> In message <19990816205639474.AAA315.413 at max41127.izone.net.au>, John
> Machin <sjmachin at lexicon.net> writes
> >On 14 Aug 99, at 19:12, Robin Becker wrote:
> >
> >> I once described perl as being like awk on
> >> steroids and lsd
> >
> >Actually perl is more like SNOBOL4 on sedatives.
> >
> I still have my snobol4 manual. All sorts of memories with that.
> >There is nothing new under the sun. SNOBOL4, which was up to version 3
> >by 1969, had goodies like arrays (like Python lists, but with multiple
> >dimensions that were fixed when the array was created), tables (like
> >Python dictionaries), user-defined functions, user-defined data-types
> ...
> --
> Robin Becker

it went even further than that.  It allowed for overloading of any and all
pre-defined symbols and had a CODE function that would let you feed in "code"
that could either augment or replace the underlying runtime engine.

Basically, you could start out in SNOBOL4 and end up with something completely
different.

-- 
Perl? Python?  If we all programmed in SNOBOL4 the way g-d intended, 
we wouldn't have to worry about debates like that!
---------------------------------------------------------------------
Howard S. Modell ???                    howard.s.modell-at-boeing.com



From pj at sgi.com  Sat Aug 28 22:41:38 1999
From: pj at sgi.com (Paul Jackson)
Date: 29 Aug 1999 02:41:38 GMT
Subject: Why do we call python a scripting language?
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com> <37c726c4.2148316@news.netset.com> <8E30E89A8PaCmAnRDLM@news.dial.oleane.com>
Message-ID: <7qa6l2$4oc37@fido.engr.sgi.com>

|> I've never even looked into freeze, but my understanding is
|> that it's not trivial to set up, or am I wrong ?

If you use a vanilla enough setup, simply downloading,
unpacking, building and installing a current Python
on a system that doesn't have other old Python fragments
laying around, with a healthy set of compile and make
tools, then freeze is easy enough.

I presume (but don't know from actual experience) that if
your setup was sufficiently weird, for some value of
"sufficiently", then freeze would start getting confused.

-- 

=======================================================================
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj



From mlh at idt.ntnu.no  Tue Aug 17 18:40:14 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 18 Aug 1999 00:40:14 +0200
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com>
Message-ID: <y0jiu6edoa9.fsf@vier.idi.ntnu.no>

Skip Montanaro <skip at mojam.com> writes:

>     Ole> I've been writing an algorithm myself, but it's incredibly slow and
>     Ole> I would like something quicker (of course :) )

result = []
for i in range(len(l)):
    element = random.choice(list)
    l.remove(element)
    result.append(element)

Probably not that quick, but...

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From kernr at mail.ncifcrf.gov  Sun Aug 15 18:43:10 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Sun, 15 Aug 1999 22:43:10 GMT
Subject: More Win32 Binaries of Python Extensions Available
References: <37b49faa.9013938@news.erols.com>
Message-ID: <37b74235.21137311@news.erols.com>

On Fri, 13 Aug 1999 22:50:09 GMT, kernr at mail.ncifcrf.gov (Robert Kern)
wrote:

>Hello guys and gals,
>
>I've added some binaries of Python extension modules to my page.

This will teach me to post too fast.  They are all here:

http://starship.python.net/crew/kernr/binaries/Binaries.html

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From jblake at stamp-coin.com  Mon Aug 16 16:24:42 1999
From: jblake at stamp-coin.com (Jonathon)
Date: Mon, 16 Aug 1999 20:24:42 +0000 (UTC)
Subject: Python CGI Scripts
In-Reply-To: <37B7C061.3824206C@objectzoo.com>
Message-ID: <Pine.LNX.4.04.9908162019270.3026-100000@natal.stamp-coin.com>

On Mon, 16 Aug 1999, Rudi Angela wrote:

>Yep, have a look at Zope, a python based opensource cgi solution:

	From my POV, Zope has two major problems.

	#1:	It expects to run as a server.  This is not a good
		thing, from the POV of most IPPs and ISPs.

	#2:	The documentation is somewhat lacking.
		<< This one is easy to fix. >>	

	I suspect that the same would be true for various other
	people as well.

	A Python equivelent to the Perl CGI.pm module would be a
	very usefull thing to have/do.

	xan

	jonathon




From cgr at formulae.co.uk  Wed Aug 25 08:09:24 1999
From: cgr at formulae.co.uk (cgr at formulae.co.uk)
Date: Wed, 25 Aug 1999 12:09:24 GMT
Subject: Unique DB access
Message-ID: <7q0mdk$hovd@eGroups.com>

Dear all,  (newbie)
Can some help me?
I have a (gadfly) database with many users data.
The database tables continue info about confidential training details
of all the users. Each user has a set of training requirements in
various subject areas. He/she will want to view the various
sub-sections of their training and be able to add/modify or delete
their personal training details.
I plan to use a HTML frame structure with a tree structure on the left
hand side. When the user clicks a node on the tree, it will take him to
a python generated HTML page on the right hand side. There will be at
least 30 nodes that link to about 5 CGI python scripts with parameters
(using 'GET').

CRITERION
The user must ONLY access his/her OWN data, not anyone else's.

PROBLEMS
Some users will use the same PC at different times of the day to enter
their details.

QUESTIONS
1. How can I easily create views of the data specific for each user?
2. By clicking on a node how can I incorporate the unique userID to
send to the python script? If I use 'GET' it is visible and therefore
hackable.

3. Can I use a cookie to store the unique key for each user on the same
PC?
What if another user uses the same PC 10 mins later?
How do I force them to login? Could they use the back button to see the
previous user details?

4. Do I use cookie.py on the server side? Are there any samples of
cookie.py implementation apart from the portion in the script itself?


Ideally,I would like the implementation to be similiar to the way
egroups is setup (login and logout), except that egroup users view all
the messages not just their own.

Thanks for any help?


Colm Rafferty





From da at ski.org  Wed Aug  4 23:20:17 1999
From: da at ski.org (David Ascher)
Date: Wed, 4 Aug 1999 20:20:17 -0700 (Pacific Daylight Time)
Subject: [Image-SIG] Proper application of the buffer interface
In-Reply-To: <006501bedeee$33ef5c20$1101a8c0@bobcat>
Message-ID: <Pine.WNT.4.05.9908042019270.40-100000@david.ski.org>

On Thu, 5 Aug 1999, Mark Hammond wrote:

> Just to extend my guesswork somewhat, there is a new built-in "buffer()"
> function.  This returns a "buffer" object.  I speculate this should be used
> in preference to Python strings when you have binary data.  

Would it also make sense to make struct.pack() return a buffer object
instead of a string?

--david ascher




From pduffin at mailserver.hursley.ibm.com  Fri Aug  6 09:48:10 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 06 Aug 1999 14:48:10 +0100
Subject: Linux why not fully dynamic?
References: <1278177653-25589496@hypernet.com>
Message-ID: <37AAE79A.7A5F@mailserver.hursley.ibm.com>

Gordon McMillan wrote:
> 
> Robin Becker asks:
> 
> > Why doesn't ./configure set up for a libpython.so.1.5.2 or whatever?
> 
> Don't know. Maybe because this kind of scheme doesn't work on
> some other Unixen?
> 

Tcl manages it and AFAIK Tcl runs on the same systems as Python will.
Also Python already has support for loadable modules and backlinking
so adding support for a shared library is actually very easy.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From tim_one at email.msn.com  Sat Aug 21 22:46:59 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 21 Aug 1999 22:46:59 -0400
Subject: slice indices (RE: More random python observations from a perl programmer)
In-Reply-To: <jwbaxter-2008991128560001@otter.olympus.net>
Message-ID: <000201beec48$9beaa840$cb2d2399@tim>

[John W. Baxter]
> My reaction to slicing being "starting from...not including" was
> "OK...I'll do it that way."  It didn't (and doesn't) seem like a
> big deal.

Then you settled for coping when you could be understanding.

unlike-the-microsoft-api-this-actually-makes-sense<wink>-ly y'rs  - tim





From tchrist at mox.perl.com  Thu Aug 19 14:48:02 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 11:48:02 -0700
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net>
Message-ID: <37bc4352@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    drew_csillag at geocities.com writes:
:There are the python manuals which are excellent, available at:
:http://www.python.org/doc/ (linked to from the homepage)
:available in your choice of HTML, PostSript, PDF, Info, LaTeX and
:windows help.

The fact that they did icky windows help and not man pages says
something significant.

--tom
-- 
    "Have the appropriate amount of fun." --Larry Wall



From thantos at brimstone.mecha  Sat Aug  7 04:53:42 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Sat, 07 Aug 1999 08:53:42 GMT
Subject: lambda fun: infinite streams
References: <37ABADB1.834A664E@go2net.com>
Message-ID: <slrn7qnsod.jjr.thantos@brimstone.mecha>

On Fri, 06 Aug 1999 20:53:21 -0700, Joe Bowbeer <joeb at go2net.com> wrote:
>I was doing a little investigating with Python's lambda expressions and
>it looks like they are indeed powerful enough to create some of my
>favorite beasts: infinitely long streams [*].

Its good to see other Scheme/Fnctional-language freaks hanging around
with the Python folks, especially those with copies of SICP.

Now, if only Python had lazy evaluation and Streams as first-class
entities, my life would be complete.  My life would be decently
complete anyway if map() worked on your infinite streams as stands.
;)

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From thomas.heller at ion-tof.com  Thu Aug 19 15:41:42 1999
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 19 Aug 1999 19:41:42 GMT
Subject: Win32 Installer - new beta
References: <1278333954-16187909@hypernet.com>	<37BB1373.6686F844@aw.sgi.com> <00c501beea24$19f87c70$4500a8c0@thomasnotebook> <37BC5900.F7914DB5@aw.sgi.com>
Message-ID: <02d701beea7a$dd633a50$4500a8c0@thomasnotebook>

> Thomas Heller wrote:
> > ...
> > Everything else is contained in the exe file and never needs to be
unpacked.
> >
> > The URL is http://starship.python.net:9673/crew/theller/installer.
> >
> > Regards
> >
> > Thomas Heller
>
> Better and better, however, let me point out the following documentation
> error:
>
> The readme-thomas.txt description for the SINGLEFILE section claims:
>
>   misc:   Any other files to include. Full pathname, or name of a
> section (the results of that section will be included).  Comma
> separated. These files will be unpacked when the exe is run.
>
> In reality these misc files are not included in the exe and not unpacked
> at run time, but merely copied into the destdir.
>
You are right. I will correct this.
SINGLEFILE does only build and collect the files needed, does not pack them.

Thanks for testing.

Thomas




From jkmuir at uniserve.com  Mon Aug 23 17:27:43 1999
From: jkmuir at uniserve.com (Justin Muir)
Date: Mon, 23 Aug 1999 16:27:43 -0500
Subject: Tkinter and classes
References: <37C0349A.EF5EEA6D@uniserve.com> <7pqtok$lp3@cs.vu.nl>
Message-ID: <37C1BCCF.A33507E4@uniserve.com>

Thanks for the responses!  They were timely and informative.

Again, thank you !


Justin.




From arcege at shore.net  Fri Aug  6 10:44:33 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Fri, 06 Aug 1999 14:44:33 GMT
Subject: embedding
References: <7nstgb$st3$1@nnrp1.deja.com>
Message-ID: <lxCq3.706$EG4.94343@news.shore.net>

vysteric at my-deja.com wrote:
: I've been reading through my docs for hours, and I can't seem to find an
: answer to this.. How do you create an object in the main namespace and
: how do you get the value of an object in the main namespace? Without
: using PyRun*

: thanks

: Patrick

Hi Patrick,

I'm surprised that noone has answered this before now.  You will want
to first import the module, then access attributes of it.

  PyObject *module__main__;
  PyObject *MyName, *whom;

  module__main__  = PyImport_ImportModule("__main__");
  /* try:...except:...*/
  if (module__main__ == NULL)
    return NULL;
  MyName = PyString_FromString("Arcege");

  if (PyObject_SetAttrString(module__main__, "person_name", MyName) == -1) {
    /* an exception */
    Py_DECREF(module__main__);
    Py_DECREF(MyName);
    return NULL;
  }
  Py_DECREF(MyName);

  whom = PyObject_GetAttrString(module__main__, "person_name");
  if (whom == NULL) {
    Py_DECREF(module__main__);
    return NULL;
  }
  PyObject_Print(whom, stdout, Py_PRINT_RAW);
  Py_DECREF(module__main__);

This is roughly equivalent to:
  import __main__
  __main__.person_name = "Arcege"
  print __main__.person_name

I use this a good deal to get access to other modules in my own code.

But, if you mean: "How can I create an object in the global namespace?"
That is slightly different; for this you can use PyEval_GetGlobals().
This returns a dictionary, on which you can then use
PyDict_GetItemString() and PyDict_SetItemString().

  -Arcege




From mlh at idt.ntnu.no  Sat Aug 21 10:00:38 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 21 Aug 1999 16:00:38 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu> <Pine.SGI.4.10.9908201606400.525821-100000@grendel>
Message-ID: <y0jso5d2pyx.fsf@vier.idi.ntnu.no>

Lars Damerow <lars at pixar.com> writes:

> On 19 Aug 1999, Tom Christiansen wrote:
> 
> > No, that's not it.  Perl's hex takes a hex number and returns 
> > a decimal one; same with oct.  Python's is the reverse.  Very weird.
> > 
> >     echo 'print oct(12)' | python
> >     14
> > 
> >     echo 'print oct(12)' | perl -l
> >     10
> > 
> > See what I mean?
> 
> I don't consider that weird at all.  Take casting, for example, in any number
> of languages.  If, in C++, I say int(var), that takes var and casts it to be an
> integer.  I might say long(num) in Python, which says "Turn num into a long."
> 
> In that case, doesn't it make sense that hex(12) would take the decimal number
> 12 and return '0xc'?  That oct(153) returns '0231'?  Functions and methods are
> typically named for the action they perform, not the data they're accepting.

Indeed... I would be very distressed if sin(0.9) returned 2... Or
sqrt(2) returned 4... Or, indeed if oct(12) returned 14. Very
disturbing.

> 
> -lars

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From dkuhlman at netcom.com  Sun Aug  1 17:33:14 1999
From: dkuhlman at netcom.com (G. David Kuhlman)
Date: 1 Aug 1999 21:33:14 GMT
Subject: Q: Why is __init__.py required
Message-ID: <7o2eeq$gs2@dfw-ixnews6.ix.netcom.com>

I have looked and looked for this.  Whenever I put an empty
__init__.py in a package, I ask myself why I have to do this.  I
read Guido's article on packages and it told me that I have to put
an __init__.py in a directory in order for the directory to be
recognized as a package.  But it doesn't say why.

I'm not asking for this behavior to be changed.  I like it the way
it is.  I just want to know *why* it has to be that way.  In what
way would the Python interpreter get confused if it didn't have
__init__.py to distinguish package directories from non-package
directories?

I've looked in the standard 1.5.2 docs and the FAQ and in Guido's
package essage and have searched Deja Vue.

Can anyone point me to an explanation?

It bugs me when I don't know why.  My parents used to have to
explain things to me like why I couldn't jump off the roof of the
house.

Thanks in advance for help.  

  - Dave



From philh at vision25.demon.co.uk  Mon Aug 30 18:06:53 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 30 Aug 99 22:06:53 GMT
Subject: Grids in Tkinter
Message-ID: <936050813snz@vision25.demon.co.uk>

Can anyone point me to anywhere that documents how Grids work in
Tkinter? A short example program would be nice.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From wware-nospam at world.std.com  Mon Aug 30 16:54:29 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Mon, 30 Aug 1999 20:54:29 GMT
Subject: Physical units
Message-ID: <FHAq2t.At6@world.std.com>

I thought somebody else might find this useful...

#########################
'''mks.py - physical units for calculations

Physical quantities are represented here as dictionaries. One key
is 'coefficient', the other keys are metric units like 'm', 'sec',
'kg'. The value for each dimension key is the power to which that unit
is raised. For instance, force is in newtons, (kg m/sec^2), so the
representation for a newton is {'sec': -2, 'kg': 1, 'm': 1,
'coefficient': 1.0}.

When quantities are multiplied or divided, the powers of units are
correctly maintained. If one unit ends up being raised to a zeroth
power, that unit is discarded in the product or ratio. If there are
no units left at all, i.e. the result is dimensionless, just return
a float.

Many physical formulas apply transcendental functions to dimensionless
ratios of quantities. For instance, the voltage across a discharging
capacitor is exp(-t/(R*C)), where t, R, and C have dimensions
(seconds, ohms, and farads), but the ratio is dimensionless. So
returning a float in such cases is a desirable behavior.

There are also some handy units and conversions at the bottom of the
file.'''

coeff = 'coefficient'
UnitsMismatch = 'UnitsMismatch'

class Quantity:
    def __init__(self,stuff,units=None):
	if units == None:
	    stuff = stuff.copy()
	    c = stuff[coeff]
	    del stuff[coeff]
	    self.stuff = stuff
	else:
	    c = 1. * stuff
	    self.stuff = units.copy()
	for k in self.stuff.keys():
	    if self.stuff[k] == 0:
		del self.stuff[k]
	self.stuff[coeff] = c
    def __repr__(self):
	str = '<%g' % self.stuff[coeff]
	for k in self.stuff.keys():
	    if k != coeff:
		str = str + ' ' + k
		if self.stuff[k] != 1:
		    str = str + '^' + `self.stuff[k]`
	return str + '>'
    def __add__(self,other):
	self.unitsMatch(other)
	units = self.stuff.copy()
	del units[coeff]
	return Quantity(self.stuff[coeff] + other.stuff[coeff],
			units)
    def __sub__(self,other):
	self.unitsMatch(other)
	units = self.stuff.copy()
	del units[coeff]
	return Quantity(self.stuff[coeff] - other.stuff[coeff],
			units)
    def __mul__(self,other):
	c = 1. * self.stuff[coeff] * other.stuff[coeff]
	u = self.multUnits(other, 1)
	if u.keys() == [ ]:
	    return c
	else:
	    return Quantity(c, u)
    def __rmul__(self,other):
	a = self.stuff.copy()
	a[coeff] = other * a[coeff]
	return Quantity(a)
    def __div__(self,other):
	try:
	    c = 1. * self.stuff[coeff] / other.stuff[coeff]
	    u = self.multUnits(other, -1)
	except AttributeError:
	    c = 1. * self.stuff[coeff] / other
	    u = self.stuff.copy()
	    del u[coeff]
	if u.keys() == [ ]:
	    return c
	else:
	    return Quantity(c, u)
    def __rdiv__(self,other):
	a = self.stuff.copy()
	b = other / a[coeff]
	del a[coeff]
	for k in a.keys():
	    a[k] = -a[k]
	return Quantity(b, a)
    def multUnits(self,other,n):
	units1 = self.stuff.copy()
	units2 = other.stuff.copy()
	del units1[coeff]
	del units2[coeff]
	for k in units1.keys():
	    if k in units2.keys():
		units1[k] = units1[k] + n * units2[k]
	for k in units2.keys():
	    if k not in units1.keys():
		units1[k] = n * units2[k]
	for k in units1.keys():
	    if units1[k] == 0:
		del units1[k]
	return units1
    def unitsMatch(self,other):
	otherkeys = other.stuff.keys()
	for k in self.stuff.keys():
	    if k not in otherkeys:
		raise UnitsMismatch
	    if k != coeff and self.stuff[k] != other.stuff[k]:
		raise UnitsMismatch

# Lotsa good stuff on units and measures at:
# http://aurora.rg.iupui.edu/UCUM/UCUM-tab.html

### Fundamental units
meter = Quantity(1,{'m':1})
kilogram = Quantity(1,{'kg':1})
second = Quantity(1,{'sec':1})
coulomb = Quantity(1,{'C':1})

Kilo = 1.e3
Mega = 1.e6
Giga = 1.e9
Tera = 1.e12
Centi = 1.e-2
Milli = 1.e-3
Micro = 1.e-6
Nano = 1.e-9
Pico = 1.e-12
Femto = 1.e-15
Atto = 1.e-18

### Conveniences and metric prefixes
meter2 = meter * meter
meter3 = meter2 * meter

# don't know the official metric names for these, if such exist
speed = meter / second
acceleration = speed / second
density = kilogram / meter3

liter = Milli * meter3
newton = kilogram * acceleration
joule = newton * meter
watt = joule / second
ampere = coulomb / second
volt = joule / coulomb
ohm = volt / ampere
farad = coulomb / volt
weber = volt * second
tesla = weber / meter2
pascal = newton / meter2
henry = weber / ampere

km = Kilo * meter
cm = Centi * meter
mm = Milli * meter
um = Micro * meter
nm = Nano * meter

gram = Milli * kilogram
mgram = Micro * kilogram
ugram = Nano * kilogram

msec = Milli * second
usec = Micro * second
nsec = Nano * second
psec = Pico * second
fsec = Femto * second

uF = Micro * farad
nF = Nano * farad
pF = Pico * farad

mH = Milli * henry
uH = Micro * henry
nH = Nano * henry

Kohm = Kilo * ohm
Mohm = Mega * ohm

### Some English measures
inch = .0254 * meter
foot = 12 * inch
yard = 3 * foot
furlong = 660 * foot
mile = 5280 * foot

gallon = 3.79 * liter
quart = .25 * gallon
pint = .5 * quart
cup = .5 * pint
fluid_ounce = .125 * cup
tablespoon = .5 * fluid_ounce
teaspoon = tablespoon / 3.

minute = 60 * second
hour = 60 * minute
day = 24 * hour
year = 365.25 * day
# There is some disagreement on years. There are Julian years (this is
# that), Gregorian years (365.2425), and tropical years (365.24219).

#################################
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From ivanlan at callware.com  Wed Aug 11 14:13:41 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 11 Aug 1999 18:13:41 GMT
Subject: converting Netscrape mailbox to Unix mbox format
References: <14256.35316.328575.118034@lrz.de>
		<m3wvv24wcg.fsf@atrus.jesus.cam.ac.uk> <14257.45321.128760.603105@lrz.de>
Message-ID: <37B1BD55.FA34A1CC@callware.com>

Hi All--

Eugene Leitl wrote:
> 
> Michael Hudson writes:
> 
>  > Last I looked Netscape mailbox format *was* unix mbox format (on unix
>  > anyway; can't speak for Netscape on windows) - have you tried pointing
>  > pine at them?
> 
> Yes. Both pine and VM. Both see a giant 5 MByte message.
> 
>  > Otherwise you could try the mailbox standard library module...

I remember MHonArc doing OK on Netscape mailboxes, but there were a lot
of missteps.  You might look into their code to see how they parse the
mbox into individual messages. ...

Take care,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From salzr at certco.com  Mon Aug  2 13:48:35 1999
From: salzr at certco.com (Rich Salz)
Date: Mon, 2 Aug 1999 13:48:35 -0400
Subject: C++ wrappers around the C API?
Message-ID: <rqbmgaa0f8fcq6@corp.supernews.com>

Has anyone written a set of C++ wrappers around Python's C API?
Is there any interest in this?






From xah at weborder.com  Sun Aug 29 01:50:36 1999
From: xah at weborder.com (Xah Lee)
Date: Sat, 28 Aug 1999 22:50:36 -0700
Subject: Wacky Programming Tales (Was Re: Why use Perl when we've got 
 Python?!)
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net>
Message-ID: <37C8CA2C.ECAE285C@weborder.com>


Brad Howes wrote:
>...
> ...now back to my language is bigger/longer/wider/stiffer than yours

No, my language is not bigger/longer/wider/stiffer than yours. My language just makes your girlfriend happier.

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 Me: "There are more than one way to do it."
 Her: <grin>



From andrew at starmedia.net  Mon Aug 16 10:29:06 1999
From: andrew at starmedia.net (Andrew Csillag)
Date: Mon, 16 Aug 1999 14:29:06 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au>
		<m3vham44g5.fsf@atrus.jesus.cam.ac.uk>
		<14257.54137.72005.705719@anthem.cnri.reston.va.us> <37b6f1c2.347828751@news.triode.net.au>
Message-ID: <37B82032.E7A11398@starmedia.net>

John Max Skaller wrote:
> 
> On Wed, 11 Aug 1999 15:48:09 -0400 (EDT), "Barry A. Warsaw" <bwarsaw at cnri.reston.va.us> wrote:
> 
> >At IPC7, JimH reported on some experiments he'd been doing
> ><http://www.jpython.org/jpython-talk-1.ppt> with whole program
> >analysis in the jpythonc tool.  He reported some impressive results,
> 
>         I'm very encouraged by this.
> 
> 
> John Max Skaller                ph:61-2-96600850
> mailto:skaller at maxtal.com.au       10/1 Toxteth Rd
> http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA

Also, you may want to look at how the SmallEiffel compiler does it's
type inferencing...  They did it well enough to get rid of virtual
function tables...  There's a few papers that describe what they did and
the essence of how it works and the rather amazing perfomance numbers
(and obviously the souce code is available to give you an implementation
to look at -- written in Eiffel).

http://SmallEiffel.loria.fr/papers/papers.html
-- 
"Programmers are just machines that convert coffee to executable code"
   -Unknown

"Drew Csillag" <drew_csillag at geocities.com>



From samschul at pacbell.net  Mon Aug 30 00:11:50 1999
From: samschul at pacbell.net (Sam Schulenburg)
Date: Sun, 29 Aug 1999 21:11:50 -0700
Subject: fo.flush() & fo.write() on windows
References: <37C9C25F.DE0BEEAB@rl.gov>
Message-ID: <Mwny3.524$Fm1.69486@typhoon-sf.snfc21.pbi.net>

An advantage of opining and closing the file to force flushing is, another
process can access the file. I use this to allow other users to view the log
files's that I create while runnig a test.

Sam Schulenburg

Roberts, Robert J <robert_j_roberts at rl.gov> wrote in message
news:37C9C25F.DE0BEEAB at rl.gov...
> Am I asking too much?
>
> pseudo code:
>
>     fo = __builtin__.open(foName, 'a')
>     foSize = os.stat(foName)[6]     # if foSize = 100...
>     fo.write('Some string.')             # fo.write() returns None
>     foSize = os.stat(foName)[6]    # ...foSize STILL = 100.
>     fo.flush()
>     foSize = os.stat(foName)[6]    # ...foSize STILL = 100.
>     fo.close()                                # 'Some string' appears in
> fo.
>     foSize = os.stat(foName)[6]    # foSize = foSize + 'Some string' #
> of bytes.
>
> I am running Python 1.5.2 on Windows 95. fo.write() appears to be
> buffering the write and returning None for bytes written on fo.write().
> fo.flush() appears to do nothing.
>
> ...or... am I missing something?
>
>





From s341625 at student.uq.edu.au  Sun Aug 29 20:38:55 1999
From: s341625 at student.uq.edu.au (Anthony Pfrunder)
Date: 30 Aug 1999 00:38:55 GMT
Subject: python and win32 com objects
References: <7qc8u8$9gq$1@msunews.cl.msu.edu>
Message-ID: <7qcjqv$vdt$1@bunyip.cc.uq.edu.au>

"Richard Hensh" <hensh at math.msu.edu> writes:

>I've noticed that one can build com objects in NT4 using the python,
>pythonwin, etc. Would this give one access to the win32 API? (Does that
>question even make sense?)

PythonWin already gives you access to swigged win32api functions.  There
is even a tinnie bit of documentation!

Cheers,

Anthony Pfrunder

--
Cheers,

Anthony Pfrunder
Computer Systems Engineering



From bashford at gage.scripps.edu  Tue Aug 17 13:20:00 1999
From: bashford at gage.scripps.edu (Don Bashford)
Date: 17 Aug 1999 10:20:00 -0700
Subject: Where to post job openings?
Message-ID: <roogg6miin.fsf@gage.scripps.edu>

Is there a place to post Python-related job openings?  I don't find
anyplace in the Python web sites.  Is posting to this newsgroup okay?




From tseaver at starbase.neosoft.com  Wed Aug  4 23:33:15 1999
From: tseaver at starbase.neosoft.com (Tres Seaver)
Date: 5 Aug 1999 03:33:15 GMT
Subject: building embedded classes as in C/C++
References: <37A78579.9856D16@psych.ucsb.edu> <slrn7qf4g2.3vvqed5.mmorris@finin.morrisland.com>
Message-ID: <591E99C29D31993B.4E574078628F2DF1.06AA13CF77DAA9A8@lp.airnews.net>

In article <slrn7qf4g2.3vvqed5.mmorris at finin.morrisland.com>,
Mitchell Morris <mmorris at mindspring.com> wrote:

>Of course it is ... everything is possible in Python, since it is Turing
>complete. On a different philosophical note, there are two concepts which
>can be expressed best by inheritance (according to the conventional C++
>wisdom which I'll parrot to the best of my ability).
>
>The most common notion is "is-a", modeled in C++ by public inheritance.
>For example, "class Penguin(Bird)" conceptually represents the notion that a
>penguin is a bird and thus could be expected to do anything a bird can do and
>respond to any input that other birds would respond to.
>
>The other common notion expressed by inheritance is "is-implemented-by".
>For example, "class TCPSocket(File)" conceptually means that a TCPSocket
>works exactly like a file, so we'll be stealing his methods directly instead
>of writing our own. In C++ this would likely be accomplished through
>private inheritance.
>
>Extra credit if you can think of an example that matches protected
>inheritance.

"Ooh! ooh!"  The classic usage is to model "is-implemented-by-for-my-kids-sake"
behavior:  

    AbstractFoo
         ^
		 +-------------+
		 |             |
    (protected)  (public,virtual)
		 |             |
   CommonFooImpl       |
		 ^             +
		 +-private--+OneOfManyDerivedFOos

The derived classes can exploit the AbstractFoo-ness of CommonFooImpl, but
nobody else can.

>
>Since your example classes "Object" and "Point" actually model neither 
>"is-a" nor "is-implemented-by", you may take as a first approximation that
>inheritance isn't going to work for you here.
>
>
>hoping-some-of-this-helps-ly y'rs,
>+Mitchell
>
>(Dang, now you've got me hyphenating!)

It's-catching-idn't-it-ly,

Tres.


-- 
---------------------------------------------------------------
Tres Seaver           tseaver at palladion.com       713-523-6582
Palladion Software    http://www.palladion.com



From bowman at montana.com  Fri Aug 20 23:47:48 1999
From: bowman at montana.com (bowman)
Date: Fri, 20 Aug 1999 21:47:48 -0600
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk> <37BC8D05.142E1DB0@aw.sgi.com> <37BC9CE8.67097FC6@vision.arc.nasa.gov> <37BCD6CF.B7A186BC@montana.com> <5lzozmskm3.fsf@eric.cnri.reston.va.us> <00dd01beeb11$e41a68e0$f29b12c2@secret.pythonware.com> <5lso5ese9x.fsf@eric.cnri.reston.va.us>
Message-ID: <37BE2163.B0AA3217@montana.com>

Guido van Rossum wrote:
>
> Good idea.  I've thought of this before, and somehow never got to it.
> I can't see any reason not to do it;

Speaking of counterintuitive and why not?, 

entrycget(index, 'menu')

returns the Tk string name of the menu, requiring a call to
nametowidget() before one can actually do something like

someMenu['bg'] = 'red'

Not a big deal, but it is rather unexpected to put a menu instance in
with entryconfigure() and get back something like '.8125431.8121234' .  

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From m.faassen at vet.uu.nl  Tue Aug 10 11:35:48 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 10 Aug 1999 15:35:48 GMT
Subject: looking for py/database ideas
References: <m2so5slkdj.fsf@netmeg.net>
Message-ID: <7opgsk$6d6$1@newshost.accu.uu.nl>

Les Schaffer <godzilla at netmeg.net> wrote:
[databases and Python]

I've had success using Python to talk to Access tables, through ODBC.

Take a look at this page:

http://www.python.org/topics/database/

for lots of info on how to interface Python with a database. It's
not difficult at all. I just treated the Access tables as 'dumb tables'
and talked to them through SQL.

If a web based interface is acceptable, you should indeed take a look at Zope.
Zope uses its own database adapters to talk to databases (through SQL again).

If you need deeper integration with Access or something like that on the
Windows platform, you may want to take a look at Python's COM extensions.
That's as far as my knowledge goes; I know they exist and that they appear
to be pretty cool. The Python DB API was enough for my needs.

Good luck!

Martijn




From connolly at w3.org  Fri Aug 20 03:32:23 1999
From: connolly at w3.org (Dan Connolly)
Date: Fri, 20 Aug 1999 07:32:23 GMT
Subject: making python man pages looks hard
Message-ID: <1103_935134343@shoal>

I'm having fun watching tchrist learn python. I learned perl from Tom
back in '90 or so. Anyway... I said my piece on perl v.s python back in '96,
and I don't have much to add since then:

Subject: Re: Python, Tcl and Perl, oh my! (was Re: tcl vs. perl)
    Date: 1996/06/26 
    Newsgroups: comp.lang.perl.misc, comp.lang.tcl, comp.lang.perl.tk, comp.lang.python
http://www.egroups.com/group/python-list/11548.html?

But, having spent most of my career translating technical
documentation from one format to another, the gripe about a
lack of man pages for python got me thinking.
I happen to be a *big fan* of the python documentation[2] as is,
but I don't see any reason why it shouldn't be available as
man pages too. Or at least... I didn't.

[2] http://www.python.org/doc/

Then I looked at the source[3]

[3] http://www.python.org/ftp/python/doc/1.5.2p1/latex-1.5.2p1.tgz

The source format is LaTeX. Converting from LaTeX is hard/messy:

==============
For example, I conjecture that it is impossible to write a program that
will extract the third word from a TeX document. It would be an easy
task for 80% of the TeX documents out there -- just skip over some
formatting stuff and grab the third bunch of characters surrounded by
whitespace. But that "formatting stuff" might be a program that
generates 100 words from the hypenation dictionary. So the simple
lexical scan of the TeX source would find a word that is not third
word of the document when printed. 

This may seem like an obscure and unimportant problem, but I
assure you that the problem of converting TeX tables to FrameMaker
MIF is just as unsolvable. 

So while "programmable" document formats have the advantage that
features can be added on a per-document basis, they suffer the
disadvantage that these features cannot be recovered by the
machine and translated in an automated fashion. 

excerpted from Toward Closure on HTML 
1994/04/07
http://www.w3.org/People/Connolly/drafts/html-essay
==============

The author of one of the python doc tools seems to agree:

===============
# Why not start from LaTeX rather than HTML?
# I could hack latex2html itself to produce Texinfo instead, or fix up
# partparse.py (which already translates LaTeX to Teinfo).
#  Pros:
#   * has high-level information such as index entries, original formatting
#  Cons:
#   * those programs are complicated to read and understand
#   * those programs try to handle arbitrary LaTeX input, track catcodes,
#     and more:  I don't want to go to that effort.  HTML isn't as powerful
#     as LaTeX, so there are fewer subtleties.
#   * the result wouldn't work for arbitrary HTML documents; it would be
#     nice to eventually extend this program to HTML produced from Docbook,
#     Frame, and more.

excerpt from
# html2texi.pl -- Convert HTML documentation to Texinfo format
# Michael Ernst <mernst at cs.washington.edu>
# Time-stamp: <1999-01-12 21:34:27 mernst>
part of [3]
===============

The bewildering array of scripts, tools, and hacks used to
generate the HTML version of the python docs is frightening!
It suggests to me that *very few people* maintain the
python docs. That's good for consistency, but it's sort
of a cathedral[3] approach: there's a sharp line between
the "blessed" modules and Everything Else.

[3] http://www.tuxedo.org/~esr/writings/cathedral-bazaar/

It would be fairly easy to convert the HTML to nroff... I think there
are tools that do that... rosettaman or something? Yes... it
seems to have an option to convert back to roff format.
http://thsun1.jinr.dubna.su/cgi-bin/rman.pl?topic=rman&section=all
ftp://ftp.cs.berkeley.edu:/ucb/people/phelps/tcltk/rman.tar.Z

The trick would be dividing up the sections. The python HTML docs
aren't self-contained like the perl man pages.

I took a quick look at the python doc-sig, but I didn't find much
relevant info... they seem to be focussed on a javadoc
work-alike... hmm... maybe that is relevant; is the python
library reference source expected to move into docstrings?
That would be cool.

Anyway... I had hoped to contribute more, but this looks harder
than I expected, and I'm done for the day.

parting shot: CPAN is cool, but I find it frustrating that I can't
read the documentation for a module without downloading
and upacking the module. for example, I can browse
the list of modules,

http://www.perl.com/CPAN/modules/00modlist.long.html

but say I find one I'm interested in:

Mac::Comm::
::OT_PPP        RdpO Control Open Transport PPP / Remote Access   CNANDOR

the only link goes to the Author contact info. Gee thanks.

"Uniform Resource Identifiers (URIs, aka URLs) are short strings that identify
resources in the web: documents, images, downloadable files, services,
electronic mailboxes, and other resources. They make resources available under
a variety of naming schemes and access methods such as HTTP, FTP, and
Internet mail addressable in the same simple way. They reduce the tedium of "log
in to this server, then issue this magic command ..." down to a single click."
 -- http://www.w3.org/Addressing/

--
Dan Connolly
http://www.w3.org/People/Connolly/




From phd at emerald.netskate.ru  Thu Aug 26 08:52:37 1999
From: phd at emerald.netskate.ru (Oleg Broytmann)
Date: Thu, 26 Aug 1999 16:52:37 +0400 (MSD)
Subject: Documentation string
In-Reply-To: <7q3c88$k8m$1@nnrp1.deja.com>
Message-ID: <Pine.LNX.4.04.9908261652000.3447-100000@emerald.netskate.ru>

Hi!

   Indent, boy, indent!

On Thu, 26 Aug 1999, sp00fD wrote:
> def foo(self):
> """A method"""

   Should be:

def foo(self):
   """A method"""

Oleg.
---- 
    Oleg Broytmann        Netskate/Inter.Net.Ru        phd at emerald.netskate.ru
           Programmers don't die, they just GOSUB without RETURN.




From mal at lemburg.com  Sun Aug  8 16:12:49 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sun, 08 Aug 1999 22:12:49 +0200
Subject: RPM for Python 1.5.2 ?
References: <slrn7qrn3u.7fj.dyoung@baltazar.baltazar.org>
Message-ID: <37ADE4C1.194C31D6@lemburg.com>

David Young wrote:
> 
> Has anyone built an RPM for Python 1.5.2 ? I think it's sort of curious
> that RedHat hasn't already. Anyway, I'm building my own package right now,
> but I'm afraid I might be wasting my time....

Try:

	http://www.andrich.net/python/download.html

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   145 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From I.Clarke at strs.co.uk  Tue Aug 17 05:05:13 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Tue, 17 Aug 1999 10:05:13 +0100
Subject: [NOISE] [OFF-TOPIC] [BUT IMPORTANT] Re: Why use Python when we've 
 got Perl?
References: <000301bee78f$6cd39de0$f22d2399@tim> <wmc4shzadhq.fsf@balard.polytechnique.fr> <m3n1vr3bw6.fsf@atrus.jesus.cam.ac.uk> <934839764snz@vision25.demon.co.uk>
Message-ID: <37B925C8.7A6DFCCC@strs.co.uk>

> Yep, Perl and python are semantically quite close (IMO). The difference
> is Python has sensible syntax.

....and so it begins again....

Ian.



From amaranda at spider.com  Mon Aug 16 05:48:38 1999
From: amaranda at spider.com (Alex Maranda)
Date: Mon, 16 Aug 1999 10:48:38 +0100
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com> <slrn7rb1uj.cf9.jstevens@bamboo.verinet.com>
Message-ID: <37B7DE75.227F@spider.com>

John Stevens wrote:
> >We don't
>  ^^
> 
> Whose we?
Oh, that...me, myself and I of course. And I am a bot, btw. I have
dreams of stealing Timbot's source and growing bigger and better...

I agree with your POV but advocacy won't get you anywhere. We're more
different than alike as communities (Perl, Python).

Regards,
Alex



From pinard at iro.umontreal.ca  Tue Aug 31 09:55:36 1999
From: pinard at iro.umontreal.ca (=?ISO-8859-1?Q?Fran=E7ois_Pinard?=)
Date: 31 Aug 1999 09:55:36 -0400
Subject: Python speed and `pcre'
Message-ID: <oqzoz89hrb.fsf@titan.progiciels-bpi.ca>

Hi, people.  Still learning and experimenting with Python. :-)

After having translated some code (not big, but not small) from Perl to
Python, I discover it runs ten times slower.  I did not learn to profile
yet, and I am not sure it would help, after having read that profiling
gives call counts, but no elapsed time.  Is that right?

My intuition tells me that the amount of regular expression matching might
provide an explanation.  Here is my set of _hypotheses_ (I'm not sure):

* Perl keeps compiled all /REGEXP/ not using string interpolation,
* Python cache for compiled REGEXP is less virtuous than I imagined.

I would be tempted to try using, in Python, a trick like `gettext' does
to avoid retranslating in already translated message.  That is, for _each_
textual `gettext' call, the compiler #define extends some code statically
allocating of a local variable used to cache the result or the call,
and use this cached result afterwards.

My problem, and I hope you will have some good advice to give for it:-),
is that I do not see how to this legibly.  I could regroup a lot of
`re.compile' calls and use the corresponding variable afterwards, but I
would much rather keep the REGEXP expressions textual near the place they
are needed, and have some:

        global gensymed-local
        if gensymed-local is None:
	    gensymed-local = re.compile(REGEXP)
        gensymed-local.sub(...)

This would have the advantage of keeping the REGEXP where it is meaningful
in the code.  Yet, creating and maintaining all those `gensymed-local'
variables by hand might be fairly tedious, not to say some code to ensure
they are all None.  (But maybe `global' already guarantees that?  If yes,
may I count on it, that is, does Python promises it is unlikely to change?)

Of course, seeing the above clutter, the idea a macro-generator is surely
tempting, but this would be fairly heavy resorting to external mechanics
for something which should be straightforward.  If Python included a
macro-generator, it would be a nice thing...

Is there something fundamental I am missing, and some just simpler code
that would have the desired effect?  Or else, is there another approach
that could give me more speed without giving too much on legibility?
I'm surely ready to accept some slowdown going from Perl to Python, but
a slowdown factor or 10 is a bit much, in my opinion.

-- 
Fran?ois Pinard   http://www.iro.umontreal.ca/~pinard




From fdrake at acm.org  Fri Aug  6 17:30:14 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 6 Aug 1999 21:30:14 GMT
Subject: API question -- decref after adding to dictionary?
In-Reply-To: <5fCq3.701$EG4.94343@news.shore.net>
References: <rps64v$0$37nspbi$7gc@news.supernews.com>
	<379EBC8B.E36F020B@lemburg.com>
	<rpu849$1$37nspbi$7gc@news.supernews.com>
	<14239.11258.615882.396388@weyr.cnri.reston.va.us>
	<5fCq3.701$EG4.94343@news.shore.net>
Message-ID: <14251.21478.91539.974922@weyr.cnri.reston.va.us>

Michael P. Reilly writes:
 > Fred, I haven't given up on my own extention writing section, just gotten
 > delayed researching the C++ side of it.  Should I just drop it?

  Sorry for the delay in responding; I tried to actually get work done 
today!  ;-)
  I think these cover different topics; the upshot of my discussion
with Rich is that Skip is creating a database of reference count
behaviors, which I (or others) can use to improve the API reference,
not the extending & embedding.
  I still welcome your contribution!


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From alan.gauld at gssec.bt.co.uk  Fri Aug 27 12:30:56 1999
From: alan.gauld at gssec.bt.co.uk (Alan Gauld)
Date: Fri, 27 Aug 1999 17:30:56 +0100
Subject: UML
References: <3F8BA119B48AD11197470008C7A493CB98B502@HQOBGEX1> <37c2afa8.67206110@news.compuserve.com> <slrn7s69p4.jb7.wtanksle@dolphin.openprojects.net> <37C3D89C.D57E9956@abg1.siemens.de> <935618487snz@vision25.demon.co.uk>
Message-ID: <37C6BD40.4FA0640@gssec.bt.co.uk>

> > Nutshell books are reference books. I tried that one as well and was
> > disappointed too.
> 
> I didn't like _UML in a Nutshell_ either, it's the worst O'Reilly
> book I've ever read.

Its not even a good reference! It misses several bits out
(cf coverage of Activity diagrams)

I prefer Instant UML from Wrox press - but it does have 
a slightly limited view of UML, and tries to explain UML 
in terms of UML which may not suit everyone!

Fowler's book gets the best reviews so far.

Alan g.
-- 
=================================================
This post represents the views of the author 
and does not necessarily accurately represent 
the views of BT.



From fredrik at pythonware.com  Fri Aug 20 04:34:37 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 08:34:37 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <012c01beeae7$f3582e20$f29b12c2@secret.pythonware.com>

(posted and mailed)

just some random comments on tom's
random comments:

Tom Christiansen <tchrist at mox.perl.com> wrote:
> GOTCHA: (low)
>     You can't use "in" on dicts.  Instead, you must use 
> d = { "fred":"wilma", "barney":"betty" }
> if d.has_key("fred"): # legal
> if "fred" in d:         # ILLEGAL
>     I don't understand why this was done.

"in" is a sequence operator; it's supposed to loop over
a sequence object (an ordered collection). dictionaries
are mappingss, not ordered collections.  some extension
types provide collections that are both sequences and
mappings.  the archives have more info on this.

> GOTCHA: (medium)
>     If you have a function that expects two arguments, such as:
> def fn(x,y):
>     return x + y
>     and you have a tuple of the two elements:
> t = (1,2)
>     You can't just call that function:
> print fn(t)
>     Or you get a "TypeError: not enough arguments; expected 2, got 1"
>     error.  You have to use this apply() function to get around this
>     issue:
> print apply(fn,t)
> 3

and praise Guido for that one...

> GOTCHA: (medium) 
>     Assignment is a statement, not an operator.  That means that
>     people are constantly writing loops that aren't really testing
>     the exit condition, as in:
> while 1:
>     line = os.readline()
>     if not line: break
>     .....

except that experienced Python programmers
write that as:

    while 1:
        line = os.readline()
        if not line:
            break

and use advanced pattern matching mechanisms to
read that as single construct.  see the archives for
more info.

(contrary to popular belief, few lines doesn't always
mean better code...)

> GOTCHA: (low)
>     Single and double quoted strings can't cross line boundaries.
>     You need triple quotes for that!

strings (single and double quoted strings are fully equivalent)
obey the same rule as all other python code: they won't go
over multiple lines, unless you add a backslash to the end of
the line.

print "spam"\
    "egg"

works, in other words.

btw, note that Python automatically concatenates string
literals:

print (
    "spam"
    "egg"
)

> DISSIMILARITY:
>     There are no private variables in a module or a class.
>     They can always be fully-qualified and accessed from
>     without.

unless you use things like Bastion, mxProxy, et al.

and the "__" prefix can be used to get private scoping,
of course.

> GOTCHA: (medium)
>     Things that return lists can't be used where a tuple is expected.
>     A function that returns a list must be coerced into a tuple to 
>     use this, though.
> def fn: return [1,2]
> print "this %d is %d here\n" % fn()  # ILLEGAL
> print "this %d is %d here\n" % tuple(fn())
>     The illegal part points out that this is an 
> TypeError: illegal argument type for built-in operation
>     Which isn't very helpful.  
> 
> GOTCHA: (high)
>     Python has no manpages!  The horror!!!!!!!!!!!!!!!!!!!!!!
>     ENODOC

well, since I've hardly ever heard anyone else asking for them.
maybe you could help us out -- I doubt it would be a major
effort to write an "esis2man" converter (look under
Doc/tools/sgmlconv).

> GOTCHA: (low)
>     Often Python's error messages leave something to be desired.
>     I don't know whether 

agreed.  they often expose too much technical stuff (how
many python programmers know what a read-only buffer is,
for example).  but this is a quite complex issue...

> GOTCHA: (high)
>     Because everything is a reference, and there's no way to dereference
>     that reference, it turns out that there is no trivial way to copy
>     a list!  This fails:
> x = [1,2]
> y = x
>     Because you don't get a new list there, just a copy to an
>     old one.  Suggested work-arounds include

it's not a copy.  it's just another reference, under a
new name.  the target object doesn't know anything
about this...

> y = x[0:]
> y = x[:]
> y = x + []
> y = x * 1
>     This forces people to think about references, again.    
>     So much for lists being first class citizens!

I don't really understand the second sentence here,
but never mind...

> GOTCHA: (medium)
>     You can't slice dictionaries at all in Python.  In Perl, it's
>     easy to slice a hash, and just as sensible.

sensible?  what are the semantics?

> 
> GOTCHA: (high)
>     As we saw with lists, because everything is a reference, and there's
>     no way to dereference that reference, this means that again there
>     is also no built-in, intuitive way to copy a dictionary. 

maybe not intuitive, but:

x = y.copy()

is pretty clear when you see it.  as are:

from copy import copy, deepcopy

x = copy(y)

note that all the above return shallow copies.  if
that's not what you want, use deepcopy instead.

> Instead, 
>     the suggested work-around is to write a loop:
> new = {}
> for key in old.keys:
>     new[key] = old[key]
>     But this is guaranteed slower, because it's not at the C level.
>     It also shows that dictionaries aren't first class citizens in python
>     as they are in Perl:

now he does that again.  what's your definition of "first
class citizen"?  is Python objects disqualified just because
the (virtual) common class ancestor doesn't implement a
standard "copy" method?

> GOTCHA: (medium)
>     Importing a variable only gives you a local copy.  In Perl, it makes
>     two names for the same object.

as in Python.

it's the assignment statement that doesn't work as you think.
look again...

> GOTCHA: (low)
>     List objects have built-in methods, like 
> l.append(x)
>     But string objects don't.  You have to import
>     from the string module to get them, and then they're
>     functions, not methods.

this will most likely change in 1.6 (which introduces
Unicode strings, among other things).

> GOTCHA: (low)
>     You have insert and append methods for lists, but only
>     a del function.  Why is it a function not a method, and
>     why isn't it spelled out?  Apprently the fashionable way
>     to do this is now
> a[2:3] = []
>     Which of course, deletes only one element.  ARG.

del a[2]

(del can affect the active namespace, which is
why it's a statement and not a function).

there's also a remove method, of course, but that's
a slightly different thing.

> GOTCHA: (medium)
>     The expression 3/4 is 0, which is false.  In Perl, 3/4 is 0.75,
>     which is what you'd expect.

depends on who you are, as earlier threads on this topic have
shown.  on the other hand, I'm pretty sure Guido won't make
this mistake the next time he creates a new language ;-)

fwiw, my copy of perl refuses to answer, though.  maybe
it's broken?

[fredrik at brain tools]$ perl  
3/4




^D
[fredrik at brain tools]$

> GOTCHA: (low)
>     Regexes default to emacs style, not egrep style!  Gads!  This is
>     surprising to anyone except an emacs weenie.  Sigh.  And just *try*
>     to read the manpage on the differences based on passed in arguments
>     establishing the style.  There aren't any manpages!  Have a nice day.

$ lynx www.python.org/doc/current/lib/module-re.html

obviously, there are more perl weenies out there than
posix weenies...

> GOTCHA: (low)
>     An out-of-bounds list reference raises an "IndexError: list index out
>     of range" exception, but not if you use a slice to get at it!
> t = range(5)
> print t[2:17]
>       [2, 3, 4]

yup.  a slice is not a given set of elements.  quoting
the language reference:

    a[i:j] selects all items with index k such that i <=k < j. 

which is exactly what's going on here...

> GOTCHA: (medium)
>     Anything that python doesn't like, it raises an exception about.
>     There is no fail-soft.  Even non-exception issues raise exceptions.
>     It's pervasive.  K&P curse languages that force people who want
>     to open() a file to wrap everything in exception code, saying that
>     "failing to open a file is hardly exceptional".  

and accordingly, the world if full of programs
that does something like:

    fh = open(backupfile, O_APPEND|...)
    write(fh, data)
    close(fh)

    remove(originalfile)

btw, if you don't want open to raise an exception, use
os.path.exists to test for the existence (better test be-
fore than after, as we all know).

if you do this, you'll still get an exception sometimes, but
only under exceptional circumstances!

(Who's K&P?  Kuchling and Peters?)

> GOTCHA: (medium)
>     You can't just interchange tuples and lists in python the way
>     you can lists and arrays in Perl.  This is an error:
> import sys  # otherwise, no argv for you, buddy
> print "First arg %s and second %s" % sys.argv[1:3]
>     because you need
> print "First arg %s and second %s" % tuple(sys.argv[1:3])

this is changing, as more code is starting to use the
abstract API.

on the other hand, there's a reason why the string
formatting operator is picky about this.  consider this:

> print "Arguments are %s" % sys.argv[1:3]

and then think a while about how things would work
if you'd replace sys.argv[1:3] with a user defined object,
and changed the string formatting operator to accept
any sequence object.

> GOTCHA: (low)
>     sort and reverse are in-place.  This leads to verbose code, such as:
> old = [1,2,3]
> new = old
> new.reverse()
>     Likewise for sort.  

http://www.python.org/doc/FAQ.html#6.20

> GOTCHA: (low)
>     You have to compiled the definition for a  function before you may
>     compile the call it. This is ok
>
> def fn(): 
>     print "hi"
> fn()

this places something called "fn" in the current namespace (the
global one, if you run that code as is), and then refers to it.

>     But putting the fn() call first:
> fn()
> def fn(): 
>     print "hi"
>     Produces the ever helpful: 
> NameError: fn
>     error message.

and this attempts to refer to something called "fn" in the current
namespace, which doesn't exist.

> So much for autoloading.

what's autoloading gotta do with it?

here's how things really work:

    1) Python compiles the *entire* module into byte codes.
    2) Python *executes* that byte code in the module's name space

when you execute "fn()", Python looks for something
called "fn" and calls it (where fn could be a function,
a class, a bound method, a lambda expression, or a
callable object instance).

when you execute "def fn", Python sets "fn" to point
to a function object.

(yes, def's a statement, looking a lot like if and while et. al.,
and behaving a lot like assignment and import).

>     ADDENDA: It turns out that def() happens *AT RUN TIME* in python.
>     That's the root of this problem.  In Perl, a sub{} is a compile-time
>     thing.

python's a dynamic language, you know ;-)

> GOTCHA: (low)
>     When you need to make an empty copy of the same type, you write
> x = y[:0]
>     So much for being typeless. Sigh.

sorry, I don't understand that comment...

(maybe because Python's not a typeless language. objects
are strongly typed, but name don't imply types; a name can
refer to different kinds of things at different times...)

> SIMILARITY:
>     Python uses None in many places where Perl uses undef().
>     But not all.  You can't over-subscript an array and get
>     away with it.

and unfortunately,

None, spam = 1, 2

doesn't really do what you would expect...

</F>




From jwilliam at xmission.com  Tue Aug 17 01:16:47 1999
From: jwilliam at xmission.com (Jerry Williams)
Date: Mon, 16 Aug 1999 23:16:47 -0600
Subject: help printing dictionary, sorted by values.
Message-ID: <37B8F03F.654FA9DE@xmission.com>

I have searched and searched the docs and am really frustrated with
them.
All I can seem to find is bits and pieces that hint about dictionaries.
I finally found a table under mappings that listed the methods.  But
I couldn't find anything that would list each data type and what methods
are available.  What I am trying to do is take a file with a bunch of
entries that appear more than once and list the count of each occurance
in order of the highest count.  So if the file looked like:
a
a
a
b
b
c
Then is would print:
3 a
2 b
1 c
I thought a dictionary would work great, but can't seem to figure out
how to sort it by values and also get the key.
list = dict.values()
list.sort()
Then I get lost.  And the counts aren't going to be unique.
Thanks for any help.  Please point me to where the methods are listed
for
the dictionary.  And some examples of dictionaries would be helpful.



From robin at jessikat.demon.co.uk  Sat Aug 21 08:56:26 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 21 Aug 1999 13:56:26 +0100
Subject: What is a troll by the way? [offtopic]
References: <HSAATIAIVmv3Ewd1@jessikat.demon.co.uk>
 <99082113432403.12112@o112>
Message-ID: <0Hm8SIA6Hqv3EwbF@jessikat.demon.co.uk>

In article <99082113432403.12112 at o112>, Fredrik Henbjork
<frehe491 at student.liu.se> writes
...
>> ppresumably the fishing reference is a mis-pronounced form of trawl
>> although given the number of English dialects troll might be correct
>> somewhere in that sense.
>> -- 
>> Robin Becker
>
>I think you are wrong. I was an active fisher until I gave it up some years ago
>due to animal rights reasons. Trolling and trawling are totally different 
>things.
>Trawling means that you basically drag a *big* net behind a *ship* while
>trolling means that you have one or several fishingrods with baits trailing
>some tens of meters behind the boat. Trawling is industrial fishing done by
>professional fisherman while trolling is mostly done by people as a hobby.
>Trolling also makes more sense in a semantical way in the expression
>"trolling for flames" than trawling since you don't use a bait (remember
>"flamebait") when trawling.
>
I stand corrected. I am no fisher.
>--
>Fredrik Henbjork
>
>Email: frehe491 at student.liu.se
>WWW: http://o112.ryd.student.liu.se
>
>

-- 
Robin Becker



From philh at vision25.demon.co.uk  Mon Aug 30 10:52:13 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Mon, 30 Aug 99 14:52:13 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <MPG.1233dfb6145766d998968c@news.mop.no> <37C9FB39.63A98303@montana.com>
Message-ID: <936024733snz@vision25.demon.co.uk>

In article <37C9FB39.63A98303 at montana.com> bowman at montana.com "bowman" writes:
> Alexander Staubo wrote:
> > 
> > A more interesting approach would be to create a Python framework that
> > loaded such Parrot scripts and instantiated necessary objects at
> > runtime.
> 
> I don't know how far I'll take it, but my current thinking is it would
> be very convenient to have Tkinter chew through a preexisting .uil file,
> and spit out a GUI.  Definitely not platform independent, and so forth,
> but, at present, having Tkinter closely resemble C/Motif is one of my
> priorities.

When I have written the parrot Tkinter backend, you (or anyone
else) will be able to write a UIL frontend for Parrot, and will then
be able to translate UIL -> Python/Tkinter (& whatever other Parrot
backend exist).    

-- 
Phil Hunt....philh at vision25.demon.co.uk




From thantos at brimstone.mecha  Sat Aug 28 14:55:03 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Sat, 28 Aug 1999 18:55:03 GMT
Subject: Fuzzy string matching?
References: <y0j7lmiuyo2.fsf@vier.idi.ntnu.no> <19990827231353095.AAA161.428@max41051.izone.net.au>  <y0jso53udoo.fsf@vier.idi.ntnu.no>
Message-ID: <slrn7sgbut.lbd.thantos@brimstone.mecha>

On 28 Aug 1999 17:28:55 +0200, Magnus L. Hetland <mlh at idt.ntnu.no> wrote:
>Could perhaps be a bit more readable, though... (Maybe I'll work on
>that...)

Hey, um, while we're talking readability, this is where I grind my own
personal axe about variable names consisting of a single, solitary
character, isn't it?  :)

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From wyzelli at yahoo.com  Sun Aug 29 03:25:20 1999
From: wyzelli at yahoo.com (Wyzelli)
Date: Sun, 29 Aug 1999 16:55:20 +0930
Subject: Larry's and Guido's REAL sigs
References: <7q9v2u$1jk$1@nnrp1.deja.com> <MPG.1232675d6e885633989ec9@nntp.hpl.hp.com>
Message-ID: <Tf5y3.5$P%4.3645@vic.nntp.telstra.net>

Larry Rosler <lr at hpl.hp.com> wrote in message
news:MPG.1232675d6e885633989ec9 at nntp.hpl.hp.com...
> In article <7q9v2u$1jk$1 at nnrp1.deja.com> on Sun, 29 Aug 1999 00:32:30
> GMT, aaron_watters at my-deja.com <aaron_watters at my-deja.com> says...
> ...
> >    http://www.chordate.com/sigs.GIF
> >
> > enjoy!  -- Aaron Watters
>
> Can't.  No such page.  Even with lower-case .gif.  Try again?
>

Works for me...

Wyzelli





From volker.hetzer at abg1.siemens.de  Wed Aug 25 07:38:37 1999
From: volker.hetzer at abg1.siemens.de (Volker Hetzer)
Date: Wed, 25 Aug 1999 13:38:37 +0200
Subject: Python 1.5.2 on HPUX 10.20
Message-ID: <37C3D5BD.72024971@abg1.siemens.de>

Hi!
This is my first look into python. I tried to install it
% ./configure --prefix=/opt/python
% make all install

and to test it
% make test

The fcntl test failed (twice):

% /opt/python/bin/python Lib/test/test_fcntl.py
Status from fnctl with O_NONBLOCK:  0
struct.pack:  '\000\002\000\000\000\000\000\000\000\000\000\000\000\000\000\000'
Traceback (innermost last):
  File "Lib/test/test_fcntl.py", line 30, in ?
    rv = fcntl.fcntl(f.fileno(), FCNTL.F_SETLKW, lockdata)
IOError: [Errno 22] Invalid argument
%

Is this serious? What am I to do?

Greetings!
Volker

-- 
Hi! I'm a signature virus! Copy me into your signature file to help me spread!



From skaller at maxtal.com.au  Tue Aug 24 22:31:52 1999
From: skaller at maxtal.com.au (John Skaller)
Date: Wed, 25 Aug 1999 12:31:52 +1000
Subject: Extending Python: exception handling
In-Reply-To: <14274.52504.789825.306617@dolphin.mojam.com>
References: <3.0.6.32.19990824122931.0099e860@mail.triode.net.au>
 <37bdcb30.796706964@news.triode.net.au>
 <14271.18469.757616.424717@dolphin.mojam.com>
 <3.0.6.32.19990824122931.0099e860@mail.triode.net.au>
Message-ID: <3.0.6.32.19990825123152.009a9100@mail.triode.net.au>

At 12:07 24/08/99 -0500, Skip Montanaro wrote:
>Later on, John wrote:
>
>	    try:
>		    y = sqrt(x)
>	    except:
>		    y = x.sqrt()
>
>    Here, the programmer is assuming that if the builtin sqrt function
>    defined for floats fails, the object might be a class emulating numbers,
>    and try to call a sqrt method.  But the Viper compiler might _know_ that
>    x is a class instance and that the builtin function 'sqrt' only applies
>    to floats, and report an error.
>
>    Even worse, it might _deduce_ that x must be a float, and then get
>    confused seeing x.sqrt(), which might require that x is an instance
>    (depending on whether sqrt is a method of floats or not).
>
>Then you're programming in a language other than Python.  

	I don't quite agree, and I want to explain what I mean by that.
First, there IS no 'precise language' Python. There is a CPython 1.5.2
_defined by the implementation_. Now, I believe that we both
would like to agree that there is a more general language "Python"
which is separate from the implementation -- it is described in 
a loose-ish way in the reference manual, although it too contains
far too many implementation details to be considered a specification
of an abstract language.

	I have pointed out that even Guido CANNOT extend any version
of Python, since behaviour is well defined for what we might
believe are 'erroneous' constructions. (there are some exceptions
where the manual says that the behaviour is undefined)

	The way around this logical conundrum must be to
specify that in SOME cases where an exception is thrown
by CPython 1.5.2, the program is in fact erroneous.

	In those cases, then, a change in semantics constitues
an extension: it will break no correct Python programs.

	Therefore, in cases like **:

	    try:
		    y = sqrt(x)
	    except:
		    y = x.sqrt()

where the user is clearly relying on properties of CPython 1.5.2,
the program is in fact NOT correct Python. Therefore, an extension
which changes the behaviour of this code does not break correct
Python programs -- although it _does_ break CPython 1.5.2 reliant
codes. (** actually, this isn't such a good example ..)

	In this specific case, the manual would need to say
that 'sqrt' requires a float argument, or an argument that
can be converted to a float. In that case, calling sqrt
on anything else is an error. THEREFORE I can ADD new
semantics in these cases without breaking any correct code.

>If you can
>reasonably determine that x is an instance and not a float, I'd expect the
>compiler to generate the equivalent of simply
>
>    y = x.sqrt()
>
>if you know that executing the try block will always fail.

	This is partly backwards. The way type inference works,
is that it has to look at the USAGE of something to determine
what it's type is.

	In order for that to work, I have to assume that the
input program is 'correct' Python. And I don't mean CPython 1.5.2,
I mean 'correct' in the abstract sense in which some code
is 'erroneous', and in that case the behaviour is UNDEFINED:
which is quite different to the well defined behaviour of
throwing an exception.

	As I pointed out above, there cannot be an abstract
notion of correct 'Python' without such assumptions.

> If it had been:
>
>    try:
>	y = sqrt(x)
>    except AttributeError:
>	y = x.sqrt()
>
>then you could generate just
>
>    y = sqrt(x)
>
>when x is a float, because an AttributeError can't occur.  (Python currently
>raises an AttributeError if an instance doesn't have a __float__ method.)

	Exactly my point. What CPython 1.5.2 does currently is NOT 
a specification for Python, otherwise it could not be extended,
CPython 1.5.1 and 1.6 wouldn't be "Python" and neither would
"JPython" be Python.

	My intent is that Viper be Python, that is, that it
be an extension of a reasonably large subset of Python.
It will be necessary to exclude some 'Python' programs,
just as JPython does. 

	In fact, 'Python' will be defined by the common
subset of CPython, JPython and Viper (and any other
implementations), together with some reasonable
interpretation of the language reference.

	As an example: I think a ZeroDivideError is
part of Python: the user can catch it, and expect it
to be thrown if division by zero is attempted.

	On the other hand, if the user uses an exception
to detect if an object has some type or not,
by 'trying out' some operation that might fail,
then their program is broken -- it isn't Python.
Because one might reasonably extend the semantics
so that the operation succeeds.

	Here is an actual working example:

	print (1).succ()

In CPython 1.5.2, this raises an exception.
But in Viper, it prints 2, because integers have
a 'succ' method. Here's another piece of code
I consider broken:

	assert type(type(1)) != types.ClassType

because in Viper, the types of standard objects
are classes. And I think that is reasonable,
and 'in the spirit of Python'. In fact, the community
has been demanding better unification of the type
system -- Viper provides it. 

For those that are interested, in the current version,
the 'type' of an object can be ANY object: not just a class!
When lookup fails, there is a 'last resort' lookup on the
type of the object.

The types of objects are NOT defined by Viper.
They're defined in Python by a client module!
So you can, for example, define any methods you like
for any type. Here is the current definition of
the type of an integer in the module 'py_types':

	class PyIntType:
		def succ(x): return x + 1
		def pred(x): return x - 1

The module 'py_types' is imported automatically.
The type of an integer must be called "PyIntType".

Actually: the attribute __typename__ specifies
the name of the type of an object in the module py_types.
For an integers, this is 'builtin' as 'PyIntType'
and cannot be changed.

In addition, the function 'type' is NOT builtin to Viper,
it is, instead, an ordinary python function defined
in the module 'py_types'. The user can replace that
definition with anything they want.

-------------------------------------------------------
John Skaller    email: skaller at maxtal.com.au
		http://www.maxtal.com.au/~skaller
		phone: 61-2-96600850
		snail: 10/1 Toxteth Rd, Glebe NSW 2037, Australia





From park6 at fas.harvard.edu  Fri Aug  6 10:12:00 1999
From: park6 at fas.harvard.edu (James Park)
Date: 6 Aug 1999 14:12:00 GMT
Subject: Shelve and Linux
Message-ID: <7oeqfg$2vh$1@news.fas.harvard.edu>

I'm having a bit of problem using the shelve module on linux

Here's a simple test script:
---------------------
#!/usr/bin/env python

import shelve
 
d = shelve.open("test.dbm", "c")
d["test"] = "haha"
d.close()
 
d= shelve.open("test.dbm")
print d["test"]
d.close()
-----------------------

On NT, it works fine.
on RedHat Linux 6.0 (2.2.7), it gives this error message:

Traceback (innermost last):
  File "./test.py", line 11, in ?
    d= shelve.open("test.dbm")
  File "/usr/lib/python1.5/shelve.py", line 152, in open
    return DbfilenameShelf(filename, flag)
  File "/usr/lib/python1.5/shelve.py", line 142, in __init__
    Shelf.__init__(self, anydbm.open(filename, flag))
  File "/usr/lib/python1.5/anydbm.py", line 83, in open
    raise error, "db type could not be determined"
anydbm.error: db type could not be determined


is this a problem with the way python is configured on my system or
something else?

thanks
jp



From cfelling at iae.nl  Mon Aug 23 18:26:11 1999
From: cfelling at iae.nl (Carel Fellinger)
Date: 24 Aug 1999 00:26:11 +0200
Subject: matching any number of any character
References: <wr34shrtkk1.fsf@gargoyle.cs.uchicago.edu> <7psgsr$hh$1@vvs.superst.iae.nl>
Message-ID: <7pshq3$i3$1@vvs.superst.iae.nl>

Carel Fellinger <cfelling at iae.nl> wrote:

>>>> rx = regex.compile('\(\(.\|\n|)+\)')

aiaia, typo, this                 ^^ should be \)
beter use re as Tim said, saves a lot of typos eh backslashes

-- 
groetjes, carel



From sjoshi at ingr.com  Fri Aug 13 16:50:39 1999
From: sjoshi at ingr.com (sjoshi at ingr.com)
Date: Fri, 13 Aug 1999 20:50:39 GMT
Subject: DefineDosDevice ??
Message-ID: <7p20er$fir$1@nnrp1.deja.com>

Hello
    I was just wondering if someone had used the DefineDosDevice
win32 function and could give me a couple of examples on how to use
it.

thanks
Sunit


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From da at ski.org  Wed Aug  4 14:52:20 1999
From: da at ski.org (David Ascher)
Date: Wed, 4 Aug 1999 11:52:20 -0700 (Pacific Daylight Time)
Subject: win32api -- long vs short filenames
In-Reply-To: <7oa0n0$72n@newsops.execpc.com>
Message-ID: <Pine.WNT.4.04.9908041150340.253-100000@rigoletto.ski.org>

On Wed, 4 Aug 1999, David Niergarth wrote:

> I'm trying to expand a mangled DOS 8+3 name (from WinNT4.0) into the long
> version of the name. There's a win32api function for doing the opposite, for
> example,
> 
> >>> win32api.GetShortPathName(r'c:\long_path_name\long_file_name.txt')
> 'c:\\LONG_P~1\\LONG_F~1.TXT'
> 
> But I can't figure out how to expand a mangled name into the original long
> name. Any ideas?

>>> win32api.FindFiles(r'c:\PROGRA~1')[0][-2]
'Program Files'

See doc in the .hlp file.

--david




From bwarsaw at cnri.reston.va.us  Fri Aug 13 13:45:24 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Fri, 13 Aug 1999 13:45:24 -0400 (EDT)
Subject: python-mode.el 3.105
Message-ID: <14260.22964.922624.370255@anthem.cnri.reston.va.us>

I've made a new version of python-mode.el available at
<http://www.python.org/emacs/python-mode/>.  This is version 3.105 and 
is what you currently get from the CVS snapshot.  Version 3.90 was the 
last one released at this URL, and 3.97 was the version release with
Python 1.5.2 final.

python-mode.el is a GNU Emacs major mode for editing and executing
Python programs.  It is compatible with both Emacs and XEmacs.

List of changes:

- Improvements to the way the JPython or CPython shell is
  selected. There is a new variable py-default-interpreter which
  determines the default shell to use for py-shell (C-c !). The shell
  is now determined the first time py-shell is invoked or a Python
  buffer is visited.

- py-shell now also takes an optional argument; when given it will
  prompt for additional switch to pass into the shell process (only on
  initial start up).

- py-toggle-shells now takes the symbols cpython and jpython as
  arguments, making it more usefully called from python-mode-hook.

- New command py-narrow-to-defun (C-x n d) which mimics the standard
  narrow-to-defun but understands Python classes and methods. With
  C-u, narrows to the most enclosing class.

- Removed special bindings for C-j to py-newline-and-indent. This
  makes Emacs diehards happier, but may break for people who
  explicitly bind or C-j to newline globally.

- The usual assortment of bugs fixes, and some performance improvements
  (for FSF Emacs). 

Enjoy,
-Barry

<p><a href="http://www.python.org/emacs/python-mode/">python-mode.el
3.105</a> - a mode for GNU Emacs and XEmacs for editing and executing
Python programs. (13-Aug-1999)



From thstr at serop.abb.se  Thu Aug 12 08:03:44 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Thu, 12 Aug 1999 14:03:44 +0200
Subject: Call for snippets
References: <199908020821.KAA07953@axil.hvision.nl>
Message-ID: <37B2B820.4726FAB2@serop.abb.se>

Hi

Hans Nowak wrote:
[a lot...]
> ?ou can check out the page at http://www.hvision.nl/~ivnowa/snippets/
[and even more...]

Hans, where are you?? It seems like your snippets page has vanished!! 
Is it only me or has something happened??
Do you have a new address??

don't-leave-us-this-way-ly y'rs
 -- Thomas



From mal at lemburg.com  Wed Aug  4 06:38:54 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 04 Aug 1999 12:38:54 +0200
Subject: Get 'ifconfig' information via Python
References: <87yafs8ruu.fsf@linuxcare.com> <37A81400.7474FD89@lemburg.com>
Message-ID: <37A8183E.1D8CD1A@lemburg.com>

M.-A. Lemburg wrote:
> 
> David N. Welton wrote:
> >
> > So, I think I'm getting close...
> >
> > import fcntl
> > import IN
> > import struct
> >
> > s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> >
> > res = fcntl.ioctl(s, IN.SIOCSIFHWADDR, ??????)

And one more thing: the first argument must be the file handle,
i.e. s.fileno() in this case.

Setting the hardware address is only allowed for super users,
BTW:

>>> fcntl.ioctl(s.fileno(), IN.SIOCSIFHWADDR, '')
Traceback (innermost last):
  File "<stdin>", line 1, in ?
IOError: (1, 'Operation not permitted')

--

Would be nice to have an output interface for ioctl() too. That
way we could access many interesting technical details
about the system running the program. This should be easy to
implement: instead of returning just the return code integer,
the function could return a tuple (RC, buffer) and then leave
the parsing of the buffer to the user.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   149 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From gmcm at hypernet.com  Mon Aug 30 09:37:47 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 30 Aug 1999 08:37:47 -0500
Subject: Newbie elementary question
In-Reply-To: <7qavtm$mha$1@nnrp1.deja.com>
Message-ID: <1276106073-51511497@hypernet.com>

rainer2207 at my-deja.com writes:

> Can someone please explain what "self" and
> "master" are?
> 
> eg.
> At the moment I have
> 
> class Interface:
>       def __init__(self, master):
>             ....
>             ....
> 
> root = Tk()
> inter = Interface(root)
> root.mainloop()
> 
> I'm assuming that master is the root window and
> self is the instance name of the window in the
> class. 

Correct.

> But I've also seen master=None. Can
> someone please explain this?

If you're asking about Python syntax, it's the way you specify a 
default value. If you're asking about Tk, I think dialogs run their 
own message loop, so don't require hooking into the root window's 
message loop.

- Gordon



From kaz at ashi.FootPrints.net  Sun Aug 22 13:03:05 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Sun, 22 Aug 1999 17:03:05 GMT
Subject: can some one help a newbie?
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <7pkke6$lut@dfw-ixnews6.ix.netcom.com> <935227166snz@vision25.demon.co.uk> <7pmv4j$gtt@dfw-ixnews14.ix.netcom.com> <935273107snz@vision25.demon.co.uk> <00fe01beecbd$95da5340$f29b12c2@secret.pythonware.com>
Message-ID: <slrn7s0bb5.e0r.kaz@ashi.FootPrints.net>

On Sun, 22 Aug 1999 16:44:20 GMT, Fredrik Lundh <fredrik at pythonware.com> wrote:
>a friend working for a big consultancy firm claims
>they're handing out the following questionnaire to
>anyone who happens to walk by outside their
>office:
>
>1. have you ever seen a computer: yes/no
>
>2. if you answered yes, have you seen a
>computer in real life (and not just on tv,
>in a magazine, etc): yes/no
>
>one yes and you get a job, two yes and you're
>hired as an expert consultant.

What if the one and only yes is in response to question 2? 

Let me guess, you go straight into management.



From tchrist at mox.perl.com  Thu Aug 19 19:40:58 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 16:40:58 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu> <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <37bc87fa@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Michael Hudson <mwh21 at cam.ac.uk> writes:
:Ah, have you discovered rlcompleter yet? Put this in your ~/.pythonrc:
:
:import rlcompleter
:readline.parse_and_bind('tab: complete')
:
:set PYTHONSTARTUP=~/.pythonrc and then in python type:
:
:d={}
:d.
:
:and whack tab a couple of times. This should definitely be better
:advertised; I use it all the time.

That just triggers:

    % python
    Python 1.5.1 (#1, Sep  3 1998, 22:51:17)  [GCC 2.7.2.3] on linux-i386
    Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
    Traceback (innermost last):
      File "/home/tchrist/.pythonrc", line 2, in ?
	readline.parse_and_bind('tab: complete')
    NameError: readline

Time to do a real install.    Hate redHate.

--tom
-- 
    I think I'm likely to be certified before Perl is...  :-) 
	--Larry Wall in <1995Feb12.061604.6008 at netlabs.com>



From skip at mojam.com  Mon Aug 16 01:47:52 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 16 Aug 1999 00:47:52 -0500 (CDT)
Subject: Why use Python when we've got Perl?
In-Reply-To: <7p5gu7$nlq@dfw-ixnews14.ix.netcom.com>
References: <37b4ba01@cs.colorado.edu>
	<7p49l5$uv7$1@nnrp1.deja.com>
	<37b634a2@cs.colorado.edu>
	<7p5gu7$nlq@dfw-ixnews14.ix.netcom.com>
Message-ID: <14263.42313.44665.455519@dolphin.mojam.com>

    Aahz> I've snipped the part of the post that I mostly agree with to
    Aahz> disagree somewhat with that last sentence.  Without going into
    Aahz> details, there were a couple of module maintainers who treated me
    Aahz> somewhat roughly.

It cuts both ways I guess.  I have next to no real Perl experience but when
I needed help with the Perl port of XML-RPC, Ken MacLeod was extremely
helpful.  Not to belabor the point, but Tom C. is also a welcome addition to
c.l.p when he drops by.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From davidopp at megsinet.net  Mon Aug 30 01:18:23 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Mon, 30 Aug 1999 01:18:23 -0400
Subject: But Sir, It's Wafer Thin!!!
Message-ID: <37CA141D.41285EFD@megsinet.net>

I only have a small question.  In fact you might say its wafer thin.

Q: What path declarations do I need to place into my autoexec.bat file
so that my computer will be able to run Python Scripts?

For some reason I am able to run at the interactive window in Pythonwin
but can't run any file that is a python script...I tried to run an
example that comes with the distribution called painter.py and the error
I get is this "WHOOPS!  I can't find a Tcl/Tk 8.0 installation
anywhere.  Please make sure that Tcl.Tk 8.0 is installed and that the
PATH environment variable is set to include the Tcl/bin directory
(or wherever TK80.DLL and TCL80.DLL are installed)."  The debugger tells
me this "Import Error: DLL load failed: One of the library files needed
to run this application cannot be found."    I used the find command to
locate both of these dll's and they are here C:\TclPlug\2.0\bin

I didn't do anything weird when installing Python on my Win98 machine.
Chose the 1.5.2 install downloaded and installed it (I am under the
understanding that this includes all needed Tcl software for Tkinter).
I also installed the additional 3MB win32 all.
Python and Pythonwin are located here C:\Program Files\Python\Pythonwin

Do I need just one path declaration or multiple path declarations and
how would they look?  My only other experience with this kind of stuff
was with setting up a pathway for a compiler and it was well documented
that my path declaration for the Mingw32 compiler should look just like
this....   PATH=c:\mingw32\bin;

Please help...I need to get over this little bump in the road and then I
will be on my way to complete mastery of this language...lol  Just a
little help...wafer thin...just point me in the right direction...tell
me I failed to adequately search the docs...anything...except don't send
the Spanish Inquisition.

David O.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/19990830/65e6dc01/attachment.html>

From mlh at idt.ntnu.no  Thu Aug 26 14:34:01 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 26 Aug 1999 20:34:01 +0200
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <37C5683D.91D61FD0@rubic.com> <14277.28442.44107.591412@dolphin.mojam.com>
Message-ID: <y0jd7wav1ba.fsf@vier.idi.ntnu.no>

Skip Montanaro <skip at mojam.com> writes:

>     Jeff> I think this proposal would break too much existing code, even for
>     Jeff> a Python 2 change.
[...]
> 
> How about a compromise?  Perhaps Aaron Watters' PyLint can be modified to
> bitch about functions without a return as the last statement?

Well... That would be a bit to strict, wouldn't it? As long as you
don't use the return value, there's no point in returning anything...
So what it probably *should* do, is argue when you assign the return
value of a function without a return as a last statement to a
variable... But that isn't possible, of course, since functions can be
assigned dynamically.

And - it might not even be true that it should be the last statement.
The return value might only be used in situations where the functions
returns a value earlier in its body...

So - this seems difficult to me. Only solution is a runtime error -
which would, of course, break a lot of code.

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From jpl at digex.net.nospamplease  Mon Aug 23 13:56:30 1999
From: jpl at digex.net.nospamplease (John Landahl)
Date: Mon, 23 Aug 1999 10:56:30 -0700
Subject: Advocacy (Re: PIGgies?)
In-Reply-To: <7pjpca$1a5@dfw-ixnews15.ix.netcom.com>; from Aahz Maruch on Fri, Aug 20, 1999 at 02:40:10PM +0000
References: <19990818171523.A22207@digex.net> <7pjpca$1a5@dfw-ixnews15.ix.netcom.com>
Message-ID: <19990823105630.A6250@digex.net>

On Fri, Aug 20, 1999 at 02:40:10PM +0000, Aahz Maruch wrote:
> In article <19990818171523.A22207 at digex.net>,
> John Landahl  <jpl at digex.net.nospamplease> wrote:
> >
> >Lately I've been considering forming a high profile user group in San
> >Francisco to help promote Python as a serious tool for mainstream use.
> >Much to my chagrin I discovered that there's already a (recent) group
> >called "BayPIGgies".  I later noticed that the term "PIGgies" is used
> >elsewhere for Python user groups.  Am I the only Python enthusiast that
> >has a problem with the term "PIGgies"?
> 
> I suspect that you've fallen for the common fallacy of thinking that
> "Python" refers to a species of snake; in fact, "Python" is a reference
> to Monty Python's Flying Circus, and most pythonistas make a point of
> heavily using irreverent humor.

Actually, no, I'm a big Python (Monty) fan from way back.  In an emergency
(ie. in dire need, or because I want to) I can regurgitate most standard
sketches, songs, and movie lines, unquestioningly to the delight of all
nearby.

My point is that an advocacy group, whose job it would be to reach many
people who are potentially unblessed by a nudgenudging cheeseshopping
lumberjacking sense of humor, would have a lot more success with a more
"mainstream" name.

I for one would really enjoy more opportunities to use Python out in the job
world, and advocacy would seem to be the way to make that happen.  Is there
any kind of organized advocacy effort already in progress, or have Python
people mostly been enjoying the language rather than talking about it?

- John



From kikutani at sprintmail.com  Thu Aug 26 10:03:18 1999
From: kikutani at sprintmail.com (Kikutani Makoto)
Date: 26 Aug 1999 14:03:18 GMT
Subject: PyKDE for Win ?
References: <7q1als$id9$1@ash.prod.itd.earthlink.net> <37c45fb2.185911967@midpoint> <slrn7s9uuf.q2.moritz@Gertrud.chaos.wg>
Message-ID: <7q3hf6$9a4$1@birch.prod.itd.earthlink.net>

On Thu, 26 Aug 1999 10:33:19 +0200, Moritz Moeller-Herrmann <moritz at news.uni-mannheim.de> wrote:

> As the QT widgets are cross-platform, it might not be too hard to port them to
> windows (might cause licensing problems though...)

Yes, I just want to use QT widgets from Python on Windows.

-- 
Kikutani, Makoto  kikutani at sprintmail.com




From mlh at idt.ntnu.no  Thu Aug 12 10:02:59 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 12 Aug 1999 16:02:59 +0200
Subject: Why is CGI module so complicated?
References: <37AB04A8.C3403427@NOSPAM.strs.co.uk> <cWYr3.1329$EG4.226668@news.shore.net> <37B15BF8.B7204705@strs.co.uk>
Message-ID: <y0jlnbhf65o.fsf@drue.idi.ntnu.no>

Ian Clarke <I.Clarke at strs.co.uk> writes:

> > : Can anyone explain what the point of the FieldStorage and
> > : MiniFieldStorage objects are?

> How about go with
> the method I described in my first post (where cgi fields are returned
> as the types that they actually are, rather than being encased in
> MiniFieldStorage objects).  Were more than one item is selected for a
> given field a list could be returned, and for a file upload a tuple
> could be returned containing the filename and the contents of the file -
> or if that is not considered extendable enough, a CGIFileUpload object
> could be returned containing the necessary fields.

Or just a File object... And then we could add a "name" attribute or
something to them... (Which would be empty in cases where the object
is not taken from a standard named file)... :)

> This would mean that for the vast majority of CGI fields which are just
> normal text, retrieval is very straight-forward and intuitive, just like
> accessing strings stored in a dictionary.  More complexity is only
> introduced when the user wants to do more complicated stuff (and even
> then, it is pretty simple).  As mentioned in my previous post, it would
> also allow the fields to be used as the substitutions for a
> TemplateDocument directly, which can be quite a useful trick.

Or - you could rewrite TemplateDocument to work with the CGI
classes... That would just be a matter of adding some code to
recognize whether it was a standard dictionary or not...

> If I get some spare time today I will create a small module which will
> provide this interface to CGI (although if anyone else wants to do it,
> feel free!).  It would obviously be better if (assuming I can convince
> others that this is a better way) this was integrated into the built-in
> cgi module.

Well... I suspect it can be *very* small for simple applications (that
don't use file upload, the methods values() or items())...

----- simplecgi.py -----

import cgi, UserDict

class FieldStorage(userdict.UserDict):
    def __init__(self):
        self.data = cgi.FieldStorage()
    def __getitem__(self, key): return self.data[key].value
        
------------------------

Or you *could* use the old classes, like FormContentDict that are
already defined (but deprecated).

> 
> Ian.

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From greg.ewing at compaq.com  Mon Aug 23 00:50:13 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 23 Aug 1999 16:50:13 +1200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu>
Message-ID: <37C0D305.21C0917@compaq.com>

Tom Christiansen wrote:
> Michael Hudson <mwh21 at cam.ac.uk> writes:
> :Well, no. How would you like to pass a tuple to a function?
> 
> I'd like using
> 
>     fn(t)
> 
> and then the function deciding whether it got enoguh arguments.

Personally I think it's a Good Thing that one
expression always maps onto one argument. It
saves potential confusion. Consider:

   def fred(a, b, c=None):

i.e. the third argument is optional. Now would
you like

   fred(1, (2, 3))

to mean fred(1, 2, 3) or fred(1, (2, 3), None)?

> No, that's not it.  Perl's hex takes a hex number and returns
> a decimal one; same with oct.  Python's is the reverse.  Very weird.

Different, certainly, and something a Perl programmer
needs to be made aware of. But one way isn't objectively
"weirder" than the other that I can see.

> Order is irrelevant.  You seem to be equating slicing with ranges.

The term "slice" seems to mean different things to
Perl and Python people. Perhaps we need to talk
about perl_slices and python_slices?

> That sounds like religion.  And aren't bound methods inherent
> memory leaks?

Only if you make cyclic data structures out of them,
but there are plenty of other ways to to that.

Greg



From robin at jessikat.demon.co.uk  Mon Aug 16 10:51:35 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 16 Aug 1999 15:51:35 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
 <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com>
 <37B7E1AC.FF4FF2CF@strs.co.uk> <ylso5kqa58.fsf@windlord.stanford.edu>
 <37B81135.1B598171@strs.co.uk>
Message-ID: <l$048EA3VCu3EwSL@jessikat.demon.co.uk>

In article <37B81135.1B598171 at strs.co.uk>, Ian Clarke
<I.Clarke at strs.co.uk> writes
>My apologies, this is wandering off-topic, I will keep it short.
>
>Ok, so your point is that I was inciting flames, but that is not in
>itself a bad thing.  I can live with that.
>
>> comp.lang.perl.misc is primarily for talking about Perl, not trying to
>> defend Perl to a Python advocate.  I'm sure the converse is true of
>> comp.lang.python.  Someone who starts a language advocacy flamewar and
>> then plays innocent about it isn't exactly holding the moral high ground.
>
>I never claimed to be perfect, Tom's post just angered me as he was
>accusing me of fanning the flames where as in he was contributing much
>more to the whole thing than I was.
>
>Ian.
I own up to a bit of flame fanning. With TC & perl it's just too
tempting.
-- 
Robin Becker



From billtut at microsoft.com  Thu Aug  5 19:11:43 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Thu, 5 Aug 1999 23:11:43 GMT
Subject: ADO example code failed:hlp!
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB9102C@RED-MSG-50>


> From: Hoon Yoon [mailto:hyoon at bigfoot.com]
> 
> 
> Running Christian's code, I get following error message. What 
> am I doing
> wrong? It's same code, but the constants do not contain 
> adModeReadWrite.
> Pls help!
> Thanks much in Adv,
> 
> import win32com.client
> db=win32com.client.Dispatch("ADODB.Connection")
> from win32com.client import constants

In order to access these constants you need to do the following:

adolib = win32com.client.gencache.EnsureModule(
	'{00000200-0000-0010-8000-00AA006D2EA4}', # TypeLib GUID for ADO
	win32con.LANG_NEUTRAL, # LCID
	2, # Major version #
	1, # Minor version #
	win32com.client.genpy.GeneratorProgress()) # Silent progress
indicator.
(The above assumes you have the latest MDAC installed, if you don't plesae
pick up the latest bits from 

Then instead of using win32com.client.Dispatch(), you can now do the
following:
	# Create ADO connection object
	db = adolib.Connection()
	# Create ADO recordset object
	rs = adolib.Recordset()

Bill



From wware-nospam at world.std.com  Thu Aug 19 12:09:57 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Thu, 19 Aug 1999 16:09:57 GMT
Subject: PIGgies?
References: <19990818171523.A22207@digex.net> <14268.9441.831503.31111@anthem.cnri.reston.va.us>
Message-ID: <FGpzKL.4wp@world.std.com>

bwarsaw at python.org wrote:
: As for the term PIGgies, I always called it a "PUG" until I heard
: about BayPIGgies started.  I don't know what a PUG is (other than a
: pugnose)

There is a kind of dog called a pug. It looks like a bulldog.

: and a PIG has this connotation of being swallowed whole by
: the Python.

Everybody seems to have this pig-hating fantasy of seeing the pig
eaten by the python. Aren't pigs supposed to be cute and smart and
friendly? "Babe, Pig in the City" and all that? And aren't ham and
bacon punishment enough? Maybe pigs need their own interest group
as self defense.

Anybody interested in a Boston-area PIGgy? I know a few other Python
enthusiasts around here.

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From aaron_watters at my-deja.com  Sun Aug 29 10:43:32 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Sun, 29 Aug 1999 14:43:32 GMT
Subject: advocacy: LinuxSlides.com
Message-ID: <7qbguj$o2$1@nnrp1.deja.com>

Got some cool slides talking about Python?
I just submitted some of mine to http://linuxslides.com/
and I thought they'd like some more.

FYI  -- Aaron Watters

====

no plan.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From ajung at sz-sb.de  Sat Aug 28 11:46:04 1999
From: ajung at sz-sb.de (Andreas Jung)
Date: Sat, 28 Aug 1999 15:46:04 GMT
Subject: Suitable for Large Applications?!
In-Reply-To: <37C7F91A.B37535@megsinet.net>; from David Oppenheimer on Sat, Aug 28, 1999 at 10:58:35AM -0400
References: <37C7F91A.B37535@megsinet.net>
Message-ID: <19990828174604.A7252@sz-sb.de>

On Sat, Aug 28, 1999 at 10:58:35AM -0400, David Oppenheimer wrote:
> Dear Group,
> 
> I have just started learning Python and would like to hear more
> discussion about its suitability for large multithousand line
> apllications.  On one hand I hear people discussing applications that
> have 10,000 plus lines and on the other I hear people say its really
> only good for small tasks and for modeling.  I'd hate to spend time
> getting real good at this language only to realize that its not well
> suited to large applications...

You can use Python to write large applications and frameworks. We
used Python in several projects as the main implementation
language with several thousands lines of code in several area:
web programming, CGI, databases (Oracle, Mysql, Postgres), Corba,
text processing, SGML/XML processing ...

> Do Python programs need to be converted to C or C++ and then compiled to
> be really useful AND run fast?  Or can you get away with creating a
> freestanding  executable by packaging the bytecode in freeze or another
> program like installer?

:-) depends on the task, the hardware and your algorithms.  In most cases I was
satisfied with the speed of Python and I've never rewritten parts of the code
in C or C++ - however it is possible to write critical parts in C/C++. It's
also possible to pack (freeze) the complete modules in a single binary - if
neccessary. Frozen binaries don't increase the execution speed
very much - it's a nice feature for distributing Python binaries
without the sources.



> What are the real performance issues.  When I hear someone say that
> copiling the Python code took hours and compiling equivalent C code took
> seconds, tends to get a newbie like me less enthused.  Should I take
> back my O'Reilly Python book and get a book on learning C++?!

What code ? The real python sources or the C code that has been
generated by the freeze module from Python ? Don't believe
in fairy-tales - Python is the most usable scripting language
available - Python is suitable both for rapid prototyping and
real-world applications. German users can take a look at my
article in the Linux-Magazin 6/99 "Python - a integration platform
for middleware" (only available in german).

May be the Python with you :-)
Andreas

-- 
                                _\\|//_
                               (' O-O ')
------------------------------ooO-(_)-Ooo--------------------------------------
   Andreas Jung, Saarbr?cker Zeitung Verlag und Druckerei GmbH
   Saarbr?cker Daten-Innovations-Center
   Untert?rkheimerstra?e 17 , D-66103 Saarbr?cken, Germany
   Phone: +49-(0)681-502-1528, Fax: +49-(0)681-502-1509
   Email: ajung at sz-sb.de (PGP key available)
-------------------------------------------------------------------------------



From mlh at idt.ntnu.no  Sun Aug 15 17:40:20 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 15 Aug 1999 23:40:20 +0200
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk> <7p74j6$km0@dfw-ixnews5.ix.netcom.com>
Message-ID: <y0jpv0oen97.fsf@vier.idi.ntnu.no>

aahz at netcom.com (Aahz Maruch) writes:

> In article <2GJsiHANGbt3Ew6i at jessikat.demon.co.uk>,
> Robin Becker  <robin at jessikat.demon.co.uk> wrote:
[...]
> 
>    x = foo()
>    y = [ foo() ]
>    z = { foo() }

Well... At least it can distinguish the first two :)

> 
> Basic Perl is in some ways easier to pick up than Python for someone who
> is a sysadmin but not a programmer.

Hm. That sounds a bit strange to me... Do you mean because it looks a
bit like awk and things like that?

> Also, while it's easier to write
> good Python programs, it's easier to debug Perl programs

Really? I don't see why... Perhaps it a personal thing? I would think
that readability is an important thing when it comes to debugging - at
least it is for me - and it seems to me that readability is a much
higher priority in Python than in Perl. (There is no dissent about
that, is there?)

> Arguing over whether Python is "better" than Perl is like arguing over
> whether European food is better than Asian.

Well - that is true, in a sense. But as long as there are no agreed
upon objective criteria to judge by, that would be true of all
discussions - no? And if there were clear criteria, the discussion
would easily be put to rest... Or perhaps this is a simplification.
Perhaps it would be more interesting in arguing over the underlying
values instead of their expression in these languages? Some people
like clarity and readability, while others like flexibility and
terseness. I suspect some people like both - in different situations.
So, maybe we should discuss which situations demand which, so that we
might become more enlightened in the matters of choosing our
programming tools to suit our needs?

advocacy-sucks-but-it-is-cool-nonetheless'ly yours - Magnus

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From greg at cosc.canterbury.ac.nz  Wed Aug 11 22:03:36 1999
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 12 Aug 1999 14:03:36 +1200
Subject: os.environ => invalid token
References: <7os33a$147t@enews4.newsguy.com>
Message-ID: <37B22B78.4A604ECF@cosc.canterbury.ac.nz>

Peter Torstenson wrote:
> 
> print os.environ['HOME']
> 
> gives : invalid token.

It should work; looks like a parsing problem. Have 
you accidentally typed a control character somewhere 
in that line? Is your indentation consistent?

-- 
Greg Ewing, Computer Science Dept, | The address below is not spam-
University of Canterbury,	   | protected, so as not to waste
Christchurch, New Zealand	   | the time of Guido van Rossum.
greg at cosc.canterbury.ac.nz



From sholden at pgrad.cs.usyd.edu.au  Sat Aug 14 00:08:05 1999
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 14 Aug 1999 04:08:05 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com>
Message-ID: <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au>

On Sat, 14 Aug 1999 03:18:19 GMT,
    John Stevens <jstevens at bamboo.verinet.com> wrote:
>On 14 Aug 1999 02:39:10 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>>On 13 Aug 1999 20:05:12 -0700,
>>	John W. Stevens <jstevens at basho.fc.hp.com> wrote:
>><snip>
>>>
>>>Thanks for the two missing magic characters (@$, and =>).  I always
>>>have to look them up.
>>
>>Learn some perl then. @ is pretty fundamental. $ is pretty fundamental.
>>What happens when you tell a scalr to be an array is pretty fundamental.
>
>Oh, I *KNOW* the semantic.  I just can't remember the syntax.
>
>As for calling the conversion of a scalar to an array "pretty
>fundamental". . . can you say "Extreme Exageration"?
>
>Tom tells me that Perl doesn't exclude non-computer scientists,
>then you guys start throwing around references to references. . .

That could simply have been a reference. Or a symbolic reference.

What is fundamental is that a @ tacked on the front indicates that it is an
array. So given @$fred, even with no knowledge of what that exactly means
you should be able to tell that it is somehow treating $fred as an array.

>
>:-)
>
>>Only if you don't know perl.
>
>No, it is confusing to the average programmer.  If it isn't
>confusing, to you, then you've spent a *LOT* of time learning
>and using Perl.
>
>>$dict is a scalar. %dict is a hash. In that
>>example a scalar was used. 
>
>Yes. . . is it a hash, or a scalar?  If it is a scalar, why
>is it called dict?  If it is a hash, then why is it prefixed
>by $?  If this is a reference instead of a scalar, then why
>doesn't it have it's own special prefix character.  ;->

It's a scalar. It is named dict because TomC called it that. It is
also named that since it is a reference to a hash. I use code like this
in C quite a bit :

Tree *tree;

Even though tree isn't a tree, but a pointer to a tree. If you don't like
that then don't use it. It's a fairly common thing to do though, I've seen
it used in pascal, C and C++, so it is not unique to perl.

>
>>At least in the discussions I have with people about python I don't
>>complain about python syntax I don't understand I just ask what it means.
>
>I am not complaining about syntax I don't understand.  I am complaining
>about syntax that is unecesarily difficult to learn, remember, and use.
>
>>Instead of calling it confusing people who actually want to learn something
>>ask what it means, and why it is so. You obviously don't want to
>>learn but only to criticise. That's fine but it would be better to just
>>post to the python group...
>
>What, you believe that you are a mind reader?
>
>You are wrong.  I know what the syntax means.  I criticize because
>it is indeed confusing, and difficult to remember.

If you know what it means then why do you continually get it wrong
throughout this thread?

>
>>You can write Perl programs that resemble sed, or awk, or C, or Lisp, or
>>Python. This is Officially Okay in Perl culture.
>
>You cannot write Perl that resemble Python.  You are required to 
>use curly braces as block delimiters.

In fact you are not. You are wrong yet again. Actually thinking about how
perl works for a minute would show you that perl is in fact flexible. 
it is not a lie when people say perl doesn't force you to write a 
particular way.

Here is some code from Damian Conway from the 'Impythonating PERL' thread
in march.

package impythonate;
use Text::Tabs;
my ($active, @bracket) = (0, ('{', ';', '}') );
sub import
{
  return 1 if ($active++);
  open CODE, $0 or die "couldn't translate $0";
  my ($code, $transcode, $lastindent) = (join('',<CODE>), '');
  $code =~ s/\\\n/ /g;
  foreach (split /\n/, $code)
  {
  my ($indent,$notempty) = /\A([ \t]*)(\S?)/;
  $indent = length(expand($indent));
  $transcode .= $bracket[1+($lastindent<=>$indent)]
    if $notempty && defined $lastindent;
  $transcode .= "$_\n";
  $lastindent = $indent if $notempty;
  }
  eval $transcode and exit or die $@;
}
1;
 
And here is his sample code that is now valid perl (although anyone who uses
it for real code should be killed) :
 
#Now newlines replace colons and indentation replaces brackets:
 
use impythonate; # STILL NEED THAT ONE SEMICOLON, DAMMIT!
 
for $i (1..10)  # COMMENTS ARE OKAY
  print "$i: "
  my $isq =  \
  $i **2 # LINE CONTINUATIONS WORK AS IN PYTHON
  print " $isq\n"
 
print "done\n"

-- 
Sam

Basically, avoid comments. If your code needs a comment to be
understood, it would be better to rewrite it so it's easier to
understand.	--Rob Pike



From roy at popmail.med.nyu.edu  Mon Aug  2 14:51:57 1999
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Mon, 02 Aug 1999 14:51:57 -0400
Subject: Timing out a BaseHTTPServer read?
Message-ID: <roy-0208991452020001@qwerky.med.nyu.edu>

I've got a subclass of BaseHTTPServer, and I'd like to get the
handle_request()  method to time-out.  Any way to do that?  Lacking that,
any way to wrap the handle_request() call in some kind of time-out
wrapper?

-- 
Roy Smith <roy at popmail.med.nyu.edu>
New York University School of Medicine




From tchrist at mox.perl.com  Thu Aug 19 12:59:20 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 09:59:20 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <slrn7ro8n1.h6v.jblaine@shell2.shore.net>
Message-ID: <37bc29d8@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    jblaine at shell2.shore.net (Jeff Blaine) writes:
:>GOTCHA: (low)
:>    You can't use "in" on dicts.  Instead, you must use 
:>	d = { "fred":"wilma", "barney":"betty" }
:>	if d.has_key("fred"):	# legal
:>	if "fred" in d:	        # ILLEGAL
:>    I don't understand why this was done.
:
:I'm very curious what would you think should be legal.
:
:Would 'in' (when used on a dict) search the keys, values, or both?
:
:Each one of those seem like a bad idea to me.  You have 2 separate
:areas containing data.  Searching each of those individually 
:(has_key()...) seems right to me.

Well, it's the polymorphism thing.  If "+" is supposed to work on anything
(which I don't agree with, but that's not my decision), I don't understand
why "in" shouldn't.  I keep thinking that everything should work on
everything.  For example, this surprises me:

    x = {1:2, 3:4}
    y = {10:20, 30:40}
    z = x + y

The latter isn't legal.  But in Perl, you can merge hashes easily.

    %x = (1,2,3,4);
    %y = (10,20,30,40);
    %z = (%x, %y);

    print $z{1}
    2
    print $z{30}
    40

If you want to do it all with references, and use pretty => arrows,
it would look like this:

    $x = {1 =>2, 3 =>4};
    $y = {10 =>20, 30 =>40);
    $z = { %$x, %$y };

    print $z->{1}
    2
    print $z->{30}
    40

Of course, that's not super efficient.  And that overright duplicates,
rather than making lists of them.  Still, that's not a bad behaviour.

I can't get used to not being able to do hash slices, as in

    @revhash{values %hash} = keys %hash;

Or 

    print sort(@hash{@desired})


But I surely see your point about not knowing whether it's a key or
a value.  The only sane thing would be the key, but that's what has_key
does.  I say "sane" because otherwise you'd need to search all the values,
and as Larry Wall says, "Doing linear scans over an associative array is
like trying to club someone to death with a loaded Uzi." (Perl Cookbook,
Chapter 5, intro). :-)

I've always resisted adding "in" to Perl because I really want people
to get out of the habit of doing linear scans, because hashing it up
into a lookup mapping is almost always the right thing in the long
run, just not for one-offs.

:Thanks for the document.  It was very interesting reading.

You're welcome.  I got so fed up with the damned flames coming
from people who don't actually even know what they're talking about,
that I decided to make sure I wasn't one of them. :-)

--tom
-- 
Can you sum up plan 9 in layman's terms? It does everything 
    Unix does only less reliably    --Ken Thompson



From skip at mojam.com  Fri Aug 27 19:50:35 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 27 Aug 1999 18:50:35 -0500 (CDT)
Subject: Returning none
In-Reply-To: <37C70821.92E810FC@prescod.net>
References: <37C55044.8CA02530@prescod.net>
	<37C5683D.91D61FD0@rubic.com>
	<14277.28442.44107.591412@dolphin.mojam.com>
	<37C70821.92E810FC@prescod.net>
Message-ID: <14279.9024.401900.142461@dolphin.mojam.com>

    Paul> I would accept this type of compromise if Python had a way of
    Paul> integrating "lint" into the edit/compile/run cycle. Most
    Paul> developers don't remember to "pylint" six modules just because
    Paul> they changed each of them in the last hour.

The same problem existed in the days when lint functionality was separate
from the C compiler.  People just had lint targets in their Makefiles.  So
you add a pylint target to your Makefile, something like (GNU make syntax):

	PYFILES = spam.py eggs.py cheese.py
	pylint: $(PYFILES:%.py=%.pyl)

	%.pyl:%.py
		pylint.py $< | tee $@

Then all you have to do is remember to "make pylint" periodically.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From olivier.deckmyn at mail.dotcom.fr  Tue Aug 17 09:32:41 1999
From: olivier.deckmyn at mail.dotcom.fr (Olivier Deckmyn)
Date: Tue, 17 Aug 1999 15:32:41 +0200
Subject: Getting the module name from a method...
Message-ID: <7pboln$n74$1@feed.teaser.fr>

Hi !

I would like a method of my own to be able to print the module it is
declared in:

ex:
the text for the module : toto.py is :

class MyClass:
    def myMethod(self):
        print "I am in module", ????????

and then
MyClass().myMethod()
should produce :
'toto'

....
Thanx for your help !

---
Olivier Deckmyn, Paris - France
"Any sufficiently advanced technology is indistinguishable from
magic." -Arthur C. Clark





From orlov at diasoft.ru  Fri Aug  6 09:43:37 1999
From: orlov at diasoft.ru (Oleg Orlov)
Date: Fri, 6 Aug 1999 17:43:37 +0400
Subject: Python in web page
References: <37a70f28.96313821@midpoint> <7o7ta9$roe$1@m2.c2.telstra-mm.net.au>
Message-ID: <7oeopb$i1k$1@ns.mtu.ru>

> Under Microsoft Internet Explorer you can use Python as an Active
Scripting
> language.
>

But python will serialize web clients due to global locking object in
pytnon.dll.
May be i missed something?





From mal at lemburg.com  Wed Aug 18 06:19:37 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Wed, 18 Aug 1999 12:19:37 +0200
Subject: Controlling Python
References: <7p93ge$2bt$1@nnrp1.deja.com> <37B82DDA.2DE0E0A6@lemburg.com> <7pdr1h$gto$1@nnrp1.deja.com>
Message-ID: <37BA88B9.170A283A@lemburg.com>

nriedlin at my-deja.com wrote:
> 
> > You can use a callback which you can register with the
> > sys.settrace() function. A less fine-grained hook is available
> > through sys.setprofile(). See the docs for more infos.
> >
> 
> You are correct, but isn't it true, that the callback is only called
> every function call? I would like to interfere every line (and so being
> able to trace my scripts). Please correct me if I'm wrong.

Please read the docs or simply try the hints I gave you...
sys.settrace() is what you want, BTW.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   135 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From gmcm at hypernet.com  Thu Aug  5 16:49:52 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 5 Aug 1999 15:49:52 -0500
Subject: Gordons installer or imputil problem?
In-Reply-To: <002101bedf6b$21c64810$4500a8c0@thomasnotebook>
Message-ID: <1278240098-21833517@hypernet.com>

Thomas Heller wrote:

> I tried out your installer (the new beta) and am really impressed.
> IMHO the best part is parsing the binary files (dlls, pyds) for
> dependecies.
> 
> I noticed a strange problem however, and I do not know where to look
> for the solution.

[import package.mod vs from package import mod]
 
> The only difference between a.py and b.py is that in the first case
>  'import package.mod' is executed before 'from package import mod',
> while in the other case the order is different. Since I'm working
> with packages a lot, and normally use 'from package import module' I
> would stronly prefer to find a solution other than rewriting all the
> import statements.

OK, just tested imputil on a .pyz built with a package= option. No 
problem using "from package import mod".

Just poked around with modulefinder - no problem there.

AHA! It's the way the zlib gets built. When you specify a package, 
the TOC has an entry for "package.__init__" and "package.mod", as it 
should. But when it's built from chasing dependencies, it has an 
entry for "package" and "package.mod", so imputil thinks "package" is 
a module...

Hack hack hack...

archivebuilder line 139 to end of function:

  for (k, v) in mf.modules.items():
    ispkg = os.path.basename(v.__file__) == '__init__.py'
    d = os.path.dirname(v.__file__)
    if not d:
      v.__file__ = os.path.join(os.getcwd(), v.__file__)
    if ispkg:
        rslt.append(k+'.__init__', v.__file__)
    else:
        rslt.append(k, v.__file__)
  return rslt

or-wait-for-the-next-rev-of-the-beta-ly y'rs

- Gordon



From thomas at xs4all.nl  Thu Aug 19 11:48:00 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Thu, 19 Aug 1999 17:48:00 +0200
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc1339@cs.colorado.edu>
References: <37bc1339@cs.colorado.edu>
Message-ID: <19990819174800.D5227@xs4all.nl>

On Thu, Aug 19, 1999 at 08:22:49AM -0700, Tom Christiansen wrote:

> This article was posted to the Perl newsgroup, but I have duplicated
> the posting here without crossposting to avoid flame wars.

Sorry, i have to reply. I wont mention much, and i definately dont want to
start a flame war (i use perl about as often as python, but python fits my
soul, perl does not.)

> GOTCHA: (low)
>     You can't use "in" on dicts.  Instead, you must use 
> 	d = { "fred":"wilma", "barney":"betty" }
> 	if d.has_key("fred"):	# legal
> 	if "fred" in d:	        # ILLEGAL
>     I don't understand why this was done.

It's an extension of Python's setup. Not everything is done to be immediately
obvious, but everything is done to be _consistent_, and instantly obvious if
you understand python. The 'in' statement works sort of like the pseudocode:

index = 0
try:
	while 1:
		if sequence[index] == findthis:
			return index - 1
		index = index + 1
except IndexError:
	return 0

(same for the 'for' statement.)

You could of course write a dictionary wrapper that takes numeric arguments
to the __getitem__ to be index in a sorted list of keys, and other arguments
as normal key retrieval... but you'd lose the ability to index on numbers.

> GOTCHA: (low)
>     I don't understand why we have so many C operators, such as "&" and
>     "|" and "~", but we don't have "&&" or "||" or "!"  We also have "<<"
>     and ">>".  It's bizarre that the relational and "?:" are missing.
>     Python uses so much from C to make people see what they expect to see,
>     but this is missing.  Why?

&& and || are and and or.

> GOTCHA: (low)
>     There are no loop labels, and therefore "break" and "continue" are
>     only through the next level.  This encourages the proliferation of
>     spurious boolean condition variables.  It was annoying when Pascal
>     made you do the same thing.  There is no "goto", which is how C
>     works around it.  As with many things in Python, here you force the
>     user to be tangled up with exceptions just to do very simple things.
>     That's not as clear a solution.

Only if you aren't used to exceptions. Exceptions are really very simple and
nice things. Especially back when they were just strings (before my time, but
still ;-)

> DISSIMILARITY:
>     Python uses [] to create a new empty list, which you
> 	then reference with [] subscripts.
>     Perl uses [] to create a new empty array, which you
> 	then reference with [] subscripts.
>     Python uses {} to create a new empty dictionary, which you
> 	then reference with [] subscripts.
>     Perl uses {} to create a new empty hash, which you
> 	then reference with {} subscripts.

You forget tuples, which you create with (), and subscript with [].
Subscripting is _always_ done with [], and it _always_ calls the
'__getitem__' member function (if any). There being no functional difference
is what makes it so nice :)

> DISSIMILARITY:
>     A class method call gets no class name as its implicit extra
>     argument the way an instance method would.  

that's because it doesn't know what class it's going to be called with. You
can assign it to another class on the fly :)

> DISSIMILARITY:
>     In Python, there is no difference between single and double quotes:
>     both interpolate C-style escapes like \t, and neither 
>     interpolates variables.  In Perl, single quotes do neither escapes
>     nor variables, but double quotes do both.

There are 'raw' strings though, 'r"<string>"', that dont do backslashes.

> DISSIMILARITY:
>     There are no private variables in a module or a class.
>     They can always be fully-qualified and accessed from
>     without.  Perl can do this with file-scoped lexicals,
>     but still can't do so with data members (modulo obscure
>     tricks such as using a non-hash for an object).

There is some namemangling though, for datamembers/functions that start with
a single _, for hiding/overriding purposes. (those mmebers get the classname
encoded in, so it's dependant on the class it was called on.)

> GOTCHA: (high)
>     Python has no manpages!  The horror!!!!!!!!!!!!!!!!!!!!!!
>     ENODOC

Damn, i hadn't even noticed that. Can you imagine ? I guess the '__doc__'
attributes on objects are enough manpage for me. ;)

> DISSIMILARITY:
>     Python doesn't convert between strings and numbers unless you tell
>     it to.  And if you don't tell it to, you get an exception.  Therefore
>     you have no idea what these 
> 	x = y + z
> 	x = y - z
> 	x = y * z
>     are really doing.  The first would concat strings or lists, the last would
>     repeat them.  Personally, I don't care for this, because I always
>     wonder what subtracting one string or list from another would be.

It's worse. _noone_ has any idea what you're doing until you're acutally
doing it. x, y and z can be any type of object, even a type of object that's
loaded in on demand or constructed on the fly.

> GOTCHA: (high)
>     This is a big surprise: strings are not atomic (although they are
>     immutable).  They are instead sequences of characters.  This comes
>     up in strange places.  For example:	
> 	>>> for c in ("weird", "bits"):
> 	...      print c
> 	... 
> 	weird
> 	bits
> 	>>> for c in ("weird"):
> 	...      print c
> 	... 
> 	w
> 	e
> 	i
> 	r
> 	d
>     The second case autosplit the characters!  Here's another:
> 	>>> print map(None, "stuff")
> 	['s', 't', 'u', 'f', 'f']
> 	>>> print map(None, "stuff", "here")
> 	[('s', 'h'), ('t', 'e'), ('u', 'r'), ('f', 'e'), ('f', None)]
>     (Python's map None is like Perl's map $_.) 

Wait, wait, you're doing two different things here. First off, you called
'for' with different types of arguments, the first time with a tuple of
strings, the second type with just a string. As a string is a sequence
object, it can be indexed, and the for loop wil 'iter' over the characters
in it. Perhaps you meant to do:

>>> for c in ("weird",):
...     print c
... 
weird

Or maybe

>>> for c in ("weird" + "bits"):
...     print c
... 
w
e
i
r
d
b
i
t
s

But they are definately not the same :)

As for the map(), almost the same. In the first instance you provide one
sequence, so map will pass 'None' a single argument, in the second 'None'
gets passed two items, and None returns it right back. If you want the
sequences turned into lists and concatenated, use + instead of ,

> GOTCHA: (high)
>     Because everything is a reference, and there's no way to dereference
>     that reference, it turns out that there is no trivial way to copy
>     a list!  This fails:
> 	x = [1,2]
> 	y = x
>     Because you don't get a new list there, just a copy to an
>     old one.  Suggested work-arounds include
> 	y = x[0:]
> 	y = x[:]
> 	y = x + []
> 	y = x * 1
>     This forces people to think about references, again.    
>     So much for lists being first class citizens!  Compare 
>     this with Perl's
> 	@x = (1,2);
> 	@y = @x;
>     Or even with references:
> 	$x = [1,2];
> 	$y = [ @$x ]; 
>     or 
> 	@y = @$x;

>>> import copy
>>> print copy.__doc__
Generic (shallow and deep) copying operations
=============================================
<etc>

> GOTCHA: (low)
>     You have insert and append methods for lists, but only
>     a del function.  Why is it a function not a method, and
>     why isn't it spelled out?  Apprently the fashionable wayto do this is
>     now
> 	a[2:3] = []
>     Which of course, deletes only one element.  ARG.

You can del a[3], or del a[3:5]

> GOTCHA: (high)
>     Because you can't use readline() to get a number, people seem to enjoy
>     calling eval() just to get a string turned into a number:
> 	import sys
> 	str = sys.stdin.readline()
> 	num = eval(x)
>     This is scary.  Even scarier is the propensity for calling input(),
>     which auto-eval()s its input to make sure it's the right "type".
>     (Funny that a soi-disant "typeless language" should be so danged
>     picky about this.)  That means you see people write:
> 	num = input("Pick a number? ")
>     But the person can supply as an answer 0x23 or 2+9 or pow(2,4)
>     or os.system("rm -rf *").  I am stunned.

Try eval(number, {}, {}). No sys or os module to call, no other variables
what so ever. Personally, i think being able to say, say, '4**5/3%2' in any
instance where the program asks you for a number is a coolness.

> GOTCHA: (low)
>     Regexes default to emacs style, not egrep style!  Gads!  This is
>     surprising to anyone except an emacs weenie.  Sigh.  And just *try*
>     to read the manpage on the differences based on passed in arguments
>     establishing the style.  There aren't any manpages!  Have a nice day.

Use re, perl regexps.

>     And I'd jolly well like to know why I wasn't allowed to use
> 	print "int of %f is %.0f" % (2.5) * 2
>     or if needed, 
> 	print "int of %f is %.0f" % ( (2.5) * 2 )

Try print "int of %f is %.0f" % ((2.5,) * 2 )
"Yes, the trailing comma counts" :)

> QUESTION:
>     What is and what is not thread safe?

Everything is thread safe. Or rather, there will at all times be only one
thread executing the Python interpreter.

Oh, damn, i erased the pow() thing. Well, anyway, 

>>> 2**2**2
16
>>> 4**4**4
Traceback (innermost last):
  File "<stdin>", line 1, in ?
OverflowError: integer pow()
>>> 4**4**4L
13407807929942597099574024998205846127479365820592393377723561443721764030073546976801874298166903427690031858186486050853753882811946569946433649006084096L

Is that what you meant ? You missed that gotcha... no automatic conversion
between int and longint. <wink>

The worst thing about python is what it does to your programming skills. gcc
is getting tired of warning me of syntax errors due to python syntax in my C
:-)

Pyton'ly y'rs,

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From jstevens at bamboo.verinet.com  Sat Aug 14 11:02:42 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: Sat, 14 Aug 1999 15:02:42 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cec7@cs.colorado.edu> <37b6185f.2442954@news.skynet.be>
Message-ID: <slrn7rb18h.cf9.jstevens@bamboo.verinet.com>

On Sat, 14 Aug 1999 07:52:51 GMT, Bart Lateur <bart.lateur at skynet.be> wrote:
>John W. Stevens wrote:
>
>>By your reasoning, Perl is very, very inconsistent (since not every
>>data type supports '+').
>
>You're looking for the Holy Grail of orthogonality, again.

No, I'm not.  I was, in fact, pointing out to Tom that his
stance was inconsistent even within Perl.

>Python is
>designed with orthogonality in mind,

No, it was not.  Python was designed to simply and clearly communicate.

Proper use of orthogonality aids in making communications clear.

>[Cute sig snipped.]

John S.

[ "Perl.  A language that looks like a C program that has been
  victimized by line noise!"

  "Yeah!?  You think so!?  Take a look at APL some time!"

  "Oh, I have a Mac at home. . ."
]



From kistler at fnmail.com  Thu Aug 19 21:36:20 1999
From: kistler at fnmail.com (Per Kistler)
Date: Fri, 20 Aug 1999 03:36:20 +0200
Subject: Viewing contents of an object
References: <01beeaa5$c7f6db60$7f8345c6@TheBlueWizard>
Message-ID: <37BCB114.37026998@fnmail.com>

Hi 

You proably want more than: 

import Tkinter
dir(Tkinter)
...
dir(Tkinter.Tributton)
...

But that's all I know.

Cheers, Per.

The Blue Wizard wrote:
> 
> Hi Pythoneers!
> 
> I would like to be able to view the content of an arbitrary object, e.g.
> view(myObject) would print
> class: myClass
> class variables:
>     counter   = 6
>     topicoftheday = "Python is fun!"
> instance variables:
>     x = 4
>     y = 7
> 
> Thanks!
> 
> Cheers!
> 
> The Blue Wizard

-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From m.faassen at vet.uu.nl  Sun Aug 22 14:05:45 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 22 Aug 1999 18:05:45 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com>
Message-ID: <7ppe5p$aj8$1@newshost.accu.uu.nl>

Fredrik Lundh <fredrik at pythonware.com> wrote:
> Martijn Faassen <m.faassen at vet.uu.nl> wrote:
>> Oh, I don't think he's that bad. He *does* flame back when he feels people
>> are flaming him, and sometimes his finger is a bit too near the flame
>> trigger, but he's nice and reasonable to the people who are nice and 
>> reasonable to him.
>> 
>> So let's try to be nice and reasonable to him?

> you mean like saying "you're right, python is completely un-
> documented" and "python 1.4 is still the current version" and
> "you're right, the assignment statement knows how to copy
> some data types but not all", etc?  that's fine with me.  the
> customer is always right, you know ;-)

I didn't mean we had to agree with him all the time. I don't even think
he said these things. I disagree with him on various points (including
the one that all docs should be in man pages otherwise they aren't included
in the comparison).

> on the other hand, I'm not sure that would have helped.
> judging from most threads that tchrist appears in (be it
> on comp.lang.perl.misc, gnu.misc.discuss, or this group),
> things often turn into a great flame fest when he's in-
> volved, by some mysterious reason.

This is possible, but if we're all nice and polite and refuse to be drawn
into flaming, everybody will be happier. And Tom Christiansen hasn't
flamed _me_ (yet :). Yes, he's argumentative. I'm far more argumentative
than I seem to be on this group, so I know what it's like. And it's not
intended to be *nasty*, in general.

Regards,

Martijn




From gherron at aw.sgi.com  Thu Aug 19 15:20:32 1999
From: gherron at aw.sgi.com (Gary Herron)
Date: Thu, 19 Aug 1999 19:20:32 GMT
Subject: Win32 Installer - new beta
References: <1278333954-16187909@hypernet.com>
		<37BB1373.6686F844@aw.sgi.com> <00c501beea24$19f87c70$4500a8c0@thomasnotebook>
Message-ID: <37BC5900.F7914DB5@aw.sgi.com>

Thomas Heller wrote:
> ...
> Everything else is contained in the exe file and never needs to be unpacked.
> 
> The URL is http://starship.python.net:9673/crew/theller/installer.
> 
> Regards
> 
> Thomas Heller

Better and better, however, let me point out the following documentation
error:

The readme-thomas.txt description for the SINGLEFILE section claims:

  misc:   Any other files to include. Full pathname, or name of a
section (the results of that section will be included).  Comma
separated. These files will be unpacked when the exe is run.

In reality these misc files are not included in the exe and not unpacked
at run time, but merely copied into the destdir.  


-- 
Dr. Gary Herron <gherron at aw.sgi.com>
206-287-5616
Alias | Wavefront
1218 3rd Ave, Suite 800, Seattle WA 98101



From donn at u.washington.edu  Fri Aug 13 17:35:01 1999
From: donn at u.washington.edu (Donn Cave)
Date: 13 Aug 1999 21:35:01 GMT
Subject: Question about order in lists
References: <7p1v28$vcf$1@news04.btx.dtag.de>
Message-ID: <7p2325$19c8$1@nntp6.u.washington.edu>

"Thorsten Z?ller" <tzoeller at t-online.de> writes:
| Suppose you write a code as follows:
| x = []                                                # empty list
| list1 = [5, 3, 7]                                  # list containing any
| numbers
| list2 = [8, 6, 3]                                  # another list containing
| any numbers
| for i in range(len(list1)):
|     x.append(list1[i-1] + list2[i-1])       # [5+8, 3+6, 7+3]
| ...
| x                                                        # print resulting
| list
|
| I'm trying to achieve that, having two lists of the same lenght containig
| any numbers, each component of the two lists is added to his pendant of the
| other lists (i.e. the first component of list1 is added to the first
| component of list2 etc). The result is saved in another list x. But if I run
| this code, I do not get [13, 9, 19] for x, but [10, 13, 9], i.e. the
| components are in a wrong order now. Why does the order change and how can I
| avoid it?

Because you index the input lists with i-1.  Use list1[i] + list2[i],
and then I think it will work as you want.

range() naturally returns a list that starts with 0, which happens to be
where Python sequences start as well.  When you decrement that initial 0
index, -1 is like len(list1)-1, hence the last shall be first.

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From paul at prescod.net  Mon Aug  9 11:28:44 1999
From: paul at prescod.net (Paul Prescod)
Date: Mon, 9 Aug 1999 15:28:44 GMT
Subject: ANN: PIL 1.0 final
References: <008e01bedfec$7ffe7110$f29b12c2@secret.pythonware.com>
Message-ID: <37AEF3AC.D98F9311@prescod.net>

PythonWare wrote:
> 
> The Python Imaging Library (PIL) adds image processing capabilities
> to your Python interpreter. This library supports many file formats,
> and provides powerful image processing and graphics capabilities.

Congratulations! This is a Really Big Deal! PIL is an awesome piece of
work. (in several senses)

 Paul Prescod



From tdfunk at nettally.com  Thu Aug 26 14:34:11 1999
From: tdfunk at nettally.com (Tom Funk)
Date: Thu, 26 Aug 1999 14:34:11 -0400
Subject: Python article on Builder.com
References: <slrn7rpqbj.atl.kjz@kzaragoza.ne.mediaone.net>
Message-ID: <MPG.122f52adbcf97bbd989698@news.nettally.com>

In an article posted Fri, 20 Aug 1999 05:38:10 GMT,
Kris J. Zaragoza (kjz at kzaragoza.ne.mediaone.net) said:
> I thought some of you might be interested in Dan Shafer's latest piece
> on Builder.Com.  It's a nice bit of PR for Python and might raise its
> visibility a little more in the Web app development world.  You can find
> the piece at:
> 
> http://buzz.builder.com/cgi-bin/WebX?14 at 236.hs95a9MoeFg^0 at .ee7bc67/0
> 

I just hopped over to C|Net and read the article as well -- as the 
ensuing religious war (Python vs. Perl vs. TCL vs. PHP vs. blah-blah-
blah).  It was....  amusing.  

Some unanswered issues have been raised that could use some input from 
some experienced Python developers.  Also, some comments from someone
who's used both Perl and Python would be welcome (only one such reader
has chimed in to date).

For instance (off the top of my head):

 - the value of Object Orientation in a scripting language
 - ISPs supporting Python and/or Zope
 - specific examples of Python/Zope-implemented applications
 - viability/ease of database access from Python

Dan paints a nice picture of Python, but the discussion needs some input 
from some experienced Python folks (Andrew? Tim? Barry? Fred? Anyone?).

-- 
-=< tom >=-
Thomas D. Funk (tdfunk at asd-web.com)      |       "Software is the lever
Software Engineering Consultant          | Archimedes was searching for"
Advanced Systems Design, Tallahassee FL. |




From kajiyama at grad.sccs.chukyo-u.ac.jp  Fri Aug  6 10:04:17 1999
From: kajiyama at grad.sccs.chukyo-u.ac.jp (Tamito Kajiyama)
Date: 06 Aug 1999 14:04:17 GMT
Subject: Help with dictionary, please
In-Reply-To: rbl@hal.cwru.edu's message of 6 Aug 1999 12:21:27 GMT
References: <7oek07$so3$1@alexander.INS.CWRU.Edu>
Message-ID: <KAJIYAMA.99Aug6230417@pseudo.grad.sccs.chukyo-u.ac.jp>

rbl at hal.cwru.edu writes:
| 
| 	if prev1 != '' and prev2 != '':
| 	 	   key = (prev2,prev1)
| 	           dict[key] = dict[key] + 1
|    and got this error:
| 	  File "Metro:Python 1.5.1:learn2.py", line 34
|           dict[key] = dict[key] + 1
|                          ^
| 	SyntaxError: invalid token
| 
| I'm obviously missing something here!

What you do seems good to me.  I wonder if you have trouble with white
space, since the code snippet shown above contains both space and tab
characters.

White space characters in indentation make sense in Python.  To avoid
mysterious syntax error, you'd better use either space or tab
characters, and not mix them.

Regards,

--
KAJIYAMA, Tamito <kajiyama at grad.sccs.chukyo-u.ac.jp>



From mstenber at cc.Helsinki.FI  Wed Aug  4 00:53:34 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 04 Aug 1999 07:53:34 +0300
Subject: building embedded classes as in C/C++
References: <37A78579.9856D16@psych.ucsb.edu>
Message-ID: <al8907sqh81.fsf@sirppi.helsinki.fi>

Andy Beall <beall at psych.ucsb.edu> writes:
> I'd like to build an embedded structure (Python class) much like I do in
> C if possible.  When I do the following, multiple instances of my final
> class overwrite each others sub-class values!

You did not pay attention, however; you set _class_wide_ variables, global
to all instances (pos, ang). Thus, the behavior was to be expected.

Try this instead:

class Object:
	def __init__(self):
		self.pos = Point()
		self.ang = Point()

> 
> -------------------------------------------------------------------------
> class Point:
> 	x = 0
> 	y = 0
> 	z = 0
> 
> class Object:
> 	pos = Point()
> 	ang = Point()
> 
> 
> obj1 = Object()
> obj2 = Object()
> 
> obj1.ang.x = 123
> 
> print obj2.ang.x    <<-- AND I FIND IT CONTAINS OBJ1's X VALUE
> -------------------------------------------------------------------------
> 
> 
> 
> In C, this would be:
> 
> -------------------------------------------------------------------------
> typedef struct Point {
> 	float  x;
> 	float  y;
> 	float  z;
> } Point;
> 
> typedef struct Object {
> 	Point  pos;
> 	Point  ang;
> } Object;
> 
> Object obj1;
> Object obj2;
> 
> obj1.ang.x = 123;
> printf("%f", obj2.ang.x);   <<-- AND NOW THE VALUES IN OBJ1 AND OBJ2 
> 				 WILL NOT BE RELATED (OBJ2 WILL JUST 
> 				 CONTAIN GARGAGE AT THIS POINT).
> -------------------------------------------------------------------------

-- 

	"If you want to travel around the world and be invited to speak at
	a lot of different places, just write a Unix operating system."
		-- Linus Torvalds



From jstevens at basho.fc.hp.com  Fri Aug 13 23:04:03 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:04:03 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132306.RAA06831@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:06 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ce93@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :$b[ 2 ] = $c;
> :
> :> That's just fine in Perl.  It's not fine in Python, because Python
> :> won't automatically grow an array.
> :
> :'Cause it doesn't have arrays (or, at least, not built in ones).
> 
> Gosh, that's a feature.  NOT.

Perl doesn't have lists.  Python doesn't have built-in arrays.

Python, however, has an array module as part of its standard library.

I will assume that a list module is available for Perl.

I wasn't trying to compare features, I was simply pointing out
that your comparison was Apples and Oranges, and therefore at
least somewhat invalid.

John S.




From mal at lemburg.com  Sun Aug 15 04:33:19 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sun, 15 Aug 1999 10:33:19 +0200
Subject: Mac ODBC
References: <199908150028.RAA26649@mb3.mailbank.com>
Message-ID: <37B67B4F.7DCA6592@lemburg.com>

Dan wrote:
> 
> Hi M.-A. Lemburg,
> 
> >Does the driver expose the ODBC API in a way that allows you to
> >link against it ? If yes, then you could try to create a new
> >subpackage for mxODBC (see my Python Pages for more infos).
> 
> Yes it does.  I initially did not think that it did, a co-worker had
> installed the ODBC package and when I asked him he said that no source
> files were included.  Your message got me thinking and checking and yes
> there were files and libraries.
> 
> When I get this working I will let you know and make the project source
> available.

Great :-) What database are you interfacing to ? If you can, you
should link against the ODBC manager if available. This gives you
more flexibility later on and would also make the interface useable
with other ODBC software for Macs.

If you need any help, feel free to ask me. Instructions for
creating a new subpackage are included in the mxODBC docs; it's
basically very simple: you just create a new subdir, then
copy everything from the mxODBC subdir into the new dir and
start editing the files.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   138 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From gmcm at hypernet.com  Fri Aug 20 01:27:47 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 20 Aug 1999 00:27:47 -0500
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <slrn7rm1ol.s1.cjc26@tankgrrl.bridget.mindriot.net>
Message-ID: <1276999043-22443525@hypernet.com>

Cliff, or a close facsimile wrote:

> I don't know how lists are implemented in Python, but if they're
> linked lists instead of arrays, then deleting an element (once
> you've already found it) should take constant time.  But then the
> append will be O(n).

They're not linked lists. They are arrays of pointers.

- Gordon



From I.Clarke at strs.co.uk  Tue Aug  3 04:05:47 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Tue, 03 Aug 1999 09:05:47 +0100
Subject: An idea - request for comments
Message-ID: <37A6A2DB.9B11FB01@strs.co.uk>

I am presently working on a CGI based system in Python, and the
experience gave me an idea for a simple Python program, I would
appreciate any comments (such as "don't bother, it's been done" or
"sure, I have loads of spare time, I will do it and release it under the
GPL").

It is to create a Python web server which automatically parses embedded
Python in HTML (there is a Perl extension that does this too).  So, for
example, the following HTML code:

<HTML>
<HEAD>
<TITLE>
{
  n = 10
  print n*2
}
</TITLE>
</HEAD>
<BODY>
{
  print n*3
}
</BODY>
</HTML>

Would output:

Page Title: 20
-------------------------
30

Internally, operation of the system would be simple.  The HTML page with
embedded python would first be converted to a python file (by having a
print statement output the HTML between the embedded python), this could
then be compiled for efficiency as when normal files are executed.  The
web server would then parse these files when requested (using an
appropriately neat method to allow access to CGI variables and Cookies
and the like from within the embedded HTML).  This would lead to a speed
improvement in that the Python interpreter need not be launched every
time.  The efficiency could be further improved by caching the compiled
python files in memory.

Cool facilities could also be implemented which sit ontop of the base
CGI and Cookie layers to emulate a "per-user" memory.  This would allow
a dictionary to be stored only to be retrieved the next time this user
accesses the site (this would be achieved using cookies and storing
information locally with a cookie-style self destruct).  Similarly, a
"per-session" memory could be implemented, making it much easier to
transfer information (such as a username and password) from one page to
the next.

Before people start pointing it out, I know this isn't the most original
idea in the world, but if it hasn't been done for Python (and I can't
find anything similar) then it is high time that it was.

I look forward to reading your comments,

Ian.



From AKiss at GEOMETRIA.hu  Mon Aug  2 10:52:30 1999
From: AKiss at GEOMETRIA.hu (Kiss, Arpad)
Date: Mon, 2 Aug 1999 16:52:30 +0200
Subject: TclError: Can't find a usable init.tcl
Message-ID: <01BEDD07.69ADF3F0@exchange1.geometria.hu>

Hi,
Tcl8.1 and Tk8.1 are not compatible with Python 1.5. There is a patch to
solve this. You may find a link to this patch in the python-list
archive.
Arpad Kiss

Rolf Wester <WESTER at ilt.fhg.de> wrote in message
news:<37A5A1BA.77BDF2B3 at ilt.fhg.de>...
> Hi,
> 
> I installed Tcl8.1, Tk8.1 and Python1.5. Trying
> 
> tk=Tkinter.Tk()
> 
> I get the error message
> ...
> TclError: Can't find a usable init.tcl in the following directories:
> 
> 
> This probably means that Tcl wasn't installed properly.
> 
> The directory list which is searched for init.tcl is obviously empty. My
> init.tcl is in the directory /usr/local/lib/tcl8.1. How can I make
> Python to search in this directory?
> 
> I would be very appreciative for help.
> 
> Rolf
> 
> ------------------------------------------------------------
> # Rolf Wester
> # Fraunhofer Institut f. Lasertechnik
> # Steinbachstr. 15, D-52074 Aachen, Germany.
> # Tel: + 49 (0) 241 8906 0, Fax: +49 (0) 241 8906 121
> # EMail: wester at ilt.fhg.de, WWW: http://www.ilt.fhg.de
> 
> 
> 
> -- 
> http://www.python.org/mailman/listinfo/python-list
> 



From evan at tokenexchange.com  Mon Aug 30 10:15:14 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Mon, 30 Aug 1999 09:15:14 -0500
Subject: But Sir, It's Wafer Thin!!!
References: <37CA141D.41285EFD@megsinet.net>
Message-ID: <imwy3.53$2y4.56820@news-dal.corridex.com>

Robert Kern has already responded with the general Windows answer, but for
W95 and beyond, don't screw with autoexec.bat and the global path.  Just add
the TCL directory to pythonw.exe's application path.  In regedit, find key
"HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\App Paths".
Look for "pythonw.exe" (*not* "python.exe") under this key.  Either add or
edit string value "Path" to include the TCL DLL directory.  You don't need
to include your whole usual path, or any reference to %PATH, since App Paths
are simply added to the global path for invocations of the app.

David Oppenheimer <davidopp at megsinet.net> wrote in message
news:37CA141D.41285EFD at megsinet.net...
I only have a small question.  In fact you might say its wafer thin.
Q: What path declarations do I need to place into my autoexec.bat file so
that my computer will be able to run Python Scripts?
For some reason I am able to run at the interactive window in Pythonwin but
can't run any file that is a python script...I tried to run an example that
comes with the distribution called painter.py and the error I get is this
"WHOOPS!  I can't find a Tcl/Tk 8.0 installation anywhere.  Please make sure
that Tcl.Tk 8.0 is installed and that the PATH environment variable is set
to include the Tcl/bin directory
(or wherever TK80.DLL and TCL80.DLL are installed)."  The debugger tells me
this "Import Error: DLL load failed: One of the library files needed to run
this application cannot be found."    I used the find command to locate both
of these dll's and they are here C:\TclPlug\2.0\bin
I didn't do anything weird when installing Python on my Win98 machine.
Chose the 1.5.2 install downloaded and installed it (I am under the
understanding that this includes all needed Tcl software for Tkinter).  I
also installed the additional 3MB win32 all.
Python and Pythonwin are located here C:\Program Files\Python\Pythonwin
Do I need just one path declaration or multiple path declarations and how
would they look?  My only other experience with this kind of stuff was with
setting up a pathway for a compiler and it was well documented that my path
declaration for the Mingw32 compiler should look just like this....
PATH=c:\mingw32\bin;
Please help...I need to get over this little bump in the road and then I
will be on my way to complete mastery of this language...lol  Just a little
help...wafer thin...just point me in the right direction...tell me I failed
to adequately search the docs...anything...except don't send the Spanish
Inquisition.
David O.





From davidopp at megsinet.net  Thu Aug 26 11:46:15 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Thu, 26 Aug 1999 11:46:15 -0400
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com>
Message-ID: <37C56146.7709B260@megsinet.net>

Perhaps tchrist should start a group called "comp.lang.flames" ?  Anyone who
posted would be vehemently disagreed with regardless of the logic of their
posting and then they could expect a couple of e-mail bombs following any
post...lol

David O.

Fredrik Lundh wrote:

> Martijn Faassen <m.faassen at vet.uu.nl> wrote:
> > Oh, I don't think he's that bad. He *does* flame back when he feels people
> > are flaming him, and sometimes his finger is a bit too near the flame
> > trigger, but he's nice and reasonable to the people who are nice and
> > reasonable to him.
> >
> > So let's try to be nice and reasonable to him?
>
> you mean like saying "you're right, python is completely un-
> documented" and "python 1.4 is still the current version" and
> "you're right, the assignment statement knows how to copy
> some data types but not all", etc?  that's fine with me.  the
> customer is always right, you know ;-)
>
> on the other hand, I'm not sure that would have helped.
> judging from most threads that tchrist appears in (be it
> on comp.lang.perl.misc, gnu.misc.discuss, or this group),
> things often turn into a great flame fest when he's in-
> volved, by some mysterious reason.
>
> maybe we should add a note to the newsgroup FAQ?
>
> </F>




From MHammond at skippinet.com.au  Thu Aug 26 01:56:13 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 26 Aug 1999 15:56:13 +1000
Subject: Help!  Dialling on modem
References: <7pemm3$fd5@enews3.newsguy.com>
Message-ID: <7q2lde$aap$1@m2.c2.telstra-mm.net.au>

If you want to talk directly to the modem, look for the "sio" (serial IO)
package at www.python.org/windows.

Alternatively, the win32 extensions include win32ras, which can be used to
dial existing dial-up networking (ie, RAS) phone-book entries.

Mark.

Michael Spencer wrote in message <7pemm3$fd5 at enews3.newsguy.com>...
>Have looked thru docs & examples on networking. Lots of info on using
>sockets and TCP. But nothing on how how to cause the modem to dial out.
>
>Lots of good reasons for not using M$ TAPI.
>
>Can anyone tell me where to find sample code on using modem to dial a
>telephone number under Win32?






From wtanksle at dolphin.openprojects.net  Thu Aug 19 13:22:55 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Thu, 19 Aug 1999 17:22:55 GMT
Subject: More random python observations from a perl programmer 
References: <37BC6986.A66B044E@starmedia.net> <199908191634.KAA01333@jhereg.perl.com>
Message-ID: <slrn7rofbf.cj3.wtanksle@dolphin.openprojects.net>

On Thu, 19 Aug 1999 16:34:12 GMT, Tom Christiansen wrote:

[talking about a Python feature]
>That's painfully inconsistent.

Your messages are interesting and enlightening and so on...  But coming
from a Perl guru this statement is... interesting.

>--tom

-- 
-William "Billy" Tanksley



From s323140 at student.uq.edu.au  Wed Aug 18 14:04:51 1999
From: s323140 at student.uq.edu.au (Rob Hodges)
Date: 19 Aug 1999 04:04:51 +1000
Subject: waiting on a callback with PyGTK
References: <m37lmtb3sz.fsf@phaedrus.zen.xom> <37BAD241.2EDD739A@bioreason.com>
Message-ID: <m3hflxm0cc.fsf@phaedrus.zen.xom>

Mitch Chapman <chapman at bioreason.com> writes:

> Rob Hodges wrote:

> > Now here's my problem: I want the program to sit there doing
> > essentially nothing until the callback returns.

> Here's some sample code showing a solution to your problem.
> 
> BTW it might be better to route this sort of question to the
> pygtk mailing list.  (But I like the fact that you're giving
> exposure to pygtk here :)  

Thanks for the very prompt answer and the heads-up on the mailing
list!  And while I'm giving exposure to pygtk here, I'd like to say
that it's just about the most glorious thing I've ever seen.  I've
never done any GUI programming before but I was able to get a basic
interface of my choosing together within a couple of hours of
untarring pygtk, in spite of the sparse documentation.  I'm seriously
impressed.  

-Rob




From fredrik at pythonware.com  Tue Aug 24 10:43:59 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 24 Aug 1999 14:43:59 GMT
Subject: Newbie question about Re
References: <37C29893.D7B2C482@gnet.tn>
Message-ID: <00e701beee3f$1d99c680$f29b12c2@secret.pythonware.com>

MB <ritm at gnet.tn> wrote:
> I just want to split a string of n x (p characters) into a list of
> p_length strings using the Re module.

umm.  since you don't really care what's IN the string,
how about using string slicing instead?  it's not only
more pythonish, it's also faster (at least in this case):

def mysplit(s, p):
    out = []
    for i in range(0, len(s), p):
        out.append(s[i:i+p])
    return out

or, if you prefer one-liners:

    list = map(lambda i, s=s, p=p: s[i:i+p], range(0, len(s), p))

</F>

    "Some people, when confronted with a problem,
    think 'I know, I'll use regular expressions.'  Now
    they have two problems." -- jwz




From winkjj at my-deja.com  Tue Aug 31 15:06:40 1999
From: winkjj at my-deja.com (winkjj at my-deja.com)
Date: Tue, 31 Aug 1999 19:06:40 GMT
Subject: Python Test Executive
Message-ID: <7qh93m$38k$1@nnrp1.deja.com>

I am in the process of using Python as our scripting language for a
test automation environment.  I have been looking for some sort of test
executive or test management solution that could help me manage and
execute many scripts.  While there is lots of software out there that
can do this, I have not found any that interface to python, or would
interface a cleanly as I would prefer.  Python has so many features
that are really beneficial for this type of application, but I have not
been able to find a tool that would allow me to take advantage of these
features.

Has anyone done some work in this area?  I have lots of ideas on how a
Python based test executive could be implemented, but unfortunately, I
am stuggling with the build yourself vs. buy off the shelf trade offs
and really don't have the time to invest in building the tooling from
the ground up.

I have also looked at TETWare, which is free on Linux, but $10K for
WinNT.  Unfortunately, I cannot deploy on Linux (just yet, but I keep
hoping...), and we must run under Win32.  And once I spend the $10K, I
still have to build a Python binding, because I have not found one
yet.  And once thats done, there really isn't a nice GUI interface to
TET anyways.  So it kind of seems like a waste of 10K.

Any ideas?

Your suggestions are greatly appreciated....

Jeff.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From arcege at shore.net  Tue Aug 17 18:22:27 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 17 Aug 1999 22:22:27 GMT
Subject: Converting a string into an object instance
References: <37B71006.48F588A1@montana.com>
Message-ID: <Dglu3.435$mV.157066@news.shore.net>

bowman <bowman at montana.com> wrote:
: Given a string representation of an object, like you would get from
: id(), is there anyway to get back the actual object? 

The value that id() returns is defined to be the object's identity, and
is only currently implimented as the object's memory address but does not
need to be.  The value is just said to be unique to that address, which
means its use in Tkinter is good, but using the value of id() does not
give you the object back.

: As background, in Tkinter, it is possible to create a menu, add
: cascades, and then set the 'menu' attribute of a Toplevel object to that
: menu, producing a menubar. If you then create another menu containing
: the pulldown entries, you can attach this menu to the toplevel cascades
: with

:    topMenu.entryconfigure(index, menu=pullDown)

: So far, so good. However, if you then want to walk the menu tree, 
: 	
:    topMenu.entrycget(index, 'menu') 

: will return a string such as .135753368.135845912, and not a copy of the
: menu object.  This is indeed the same thing you get if you 'print
: pullDown', but is not too useful for manipulating the attributes or
: entries of the menu.

True, it isn't all that useful; the common method is to save the widget
when you want to use it later.

: It is easy enough to just refer to the menu directly, but in some
: instances it might be useful to have the ability to modify the 4th entry
: in the 3rd cascade with only a reference to the top level object.

Since the object's identity is unique, it's good to use as a hash value
(in fact, it is the default hash value for a class), so you can create
a dictionary that captures the information for you:
  widget_map = {}
  widget_map[str(widget)] = widget

But for this problem specifically, you can use my MenuEntry module:
  http://starship.python.net/crew/arcege/modules/MenuEntry.py

It lets you do something like the following:
  menu = Menu(mbutton)
  open = MenuCommand(menu, label='Open...', command=self.openfile_dialog)
  save = MenuCommand(menu, label='Save', command=self.savefile_default)
  saveas = MenuCommand(menu, label='Save...', command=self.savefile_dialog)
  MenuSeparator(menu)
  MenuCommand(menu, label='Quit', command=self.quitapp)
  save.config(state=DISABLED)  # "grey-out" the menu option
  print save['label']

MenuCascade objects create a submenu for you (called "submenu").

  -Arcege




From skaller at maxtal.com.au  Fri Aug 27 22:21:41 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sat, 28 Aug 1999 02:21:41 GMT
Subject: Python Type System: An idea for unification
Message-ID: <37c72e93.1411973891@news.triode.net.au>

While implementing Viper, I've been playing with the python type
system. Some of the results seem interesting enough to discuss,
and may be useful for C -&- J Python.

One of the things that is often requested for python is that
builtin types be class instances. Viper does _not_ do this:
AFAICS it is theoretically impossible (and  less useful than
the technique described below).

Here's what I do instead. Every object has an associated
type object. The function 'type(x)' gets the type object
associated with an object x. 

When a lookup on an object fails, lookup is done
on the type object of the object, before finally
issuing an AttributeError.

If the special lookup on the type object succeeds
and the result is a callable object, it is bound
to the original object to make a bound method,
as if the type object were a class, and the lookup
were for an instance attribute.

Now, for the builtin Viper objects, there is a 
python file 'py_types.py' which is loaded at startup
by the viperi interpreter. This file contains
the definitions of the type objects for the builtin types,
it is an ordinary python module!

What do the definitions look like? Well, we need the
type objects to support the alternate lookup mechanism,
so the obvious thing to do is use classes. Here is an
example:

	class PyIntType:
		def succ(x): return x + 1
		def pred(x): return x - 1

To use this we try out the Viper code:

	x = 1
	print x . succ()

and it prints "2". The object x is NOT a class instance,
but it's type is a class, 

	py_types.PyIntType 

and the class methods of that class can be called on
any int object. If you say:

	import types
	print types.IntType

you will get

	"<class py_types.PyIntType>"

Viper has no special Type type: if you say

	print types.TypeType

you will get

	"<class py_types.PyClassType>"

which is just an ordinary class defining the type
of classes. The potential infinite regress looking up
types is halted whenever the type of an object is itself.

It is important to note that for a PyInstance -- the kind of
object created by a class like:

	class X: pass
	x = X()

the _class_ of x is X but the _type_ of x is
py_types.PyInstanceType = types.InstanceType.

Several advantages of this scheme are immediately apparent:

	1) the type system is unified
	2) the type of an object is a USER accessible object

To see what (2) means: you can modify the methods existing
types support by simply changing the file 'py_types.py'.
Indeed, you can add new methods to a type by simply
adding them to the class at run time! [In fact,
you can change the type of an object at run time -- 
a risky technique, but possibly useful]

But even better, you can define new types, and you
can do it IN PYTHON (well, in Viper anyhow :-)
You do not need to write any C code to add a new type to the type system.
(Although this will still be necessary to interface external C functions)

Indeed -- and this is what most people want I think -- you can
use INHERITANCE to derive new types from old ones
(provided the old type is a class, of course).

Now, in Viper, the type of an object can be a class,
but it doesn't _have_ to be. It can be _any_ object,
even the object itself. Some care is needed that
if a lookup is done, it will terminate: it is safe if there
are no circularities, or the only cycle is of length 1.

What use are objects other than classes as types?
I have NOT tried this yet, but I will explain the idea
anyhow. Consider the type:

	ListOfInt

representing a list of integers: unlike the usual list type,
we want a list that can _only_ contain integers.
Now, I could add this type to Viper. Indeed, with enough
support from the system, you could do it too, in python (see above).

However, the general notion of 'list of X' for some type X
could be implemented instead. Here's the idea:

	class ListOf:
		def __init__(self,etype):
			self.etype = etype
		def __call__(self): 
			obj = new_object(self)
			obj.data = []
			return obj
		# ...

Here, we use a class ListOf as a _generic_ type for all
lists of the same kind of type, and we use an _instance_
of that class for the type: the argument to the __init__ method
specifies the type:

	ListOfInt = ListOf(IntType)

will construct the type object for a list of ints.  And here is an actual
list of ints:

	x = ListOfInt()

Here, the __call__ method defined for the generic type ListOf
is called, because ListOfInt is an instance of the class ListOf.
The 'new_object' function creates a new object, and sets the type
of the object to its argument, which is an instance of ListOf
for which the 'etype' parameter is IntType.

When an attribute lookup is done on x, it will proceed to the
type object. The type object is ListOfInt, so any attributes
of that object will be searched. Such a search automatically
looks in the class ListOf, if it fails in the instance dictionary,
because it is the class of ListOfInt. Therefore, methods
of 'ListOf' become methods of ListOfInt and thus
methods of a list of integers.

We need to check where the binding occurs. In the
lookup on ListOf, a function found there is callable,
and so is bound to the instance ListOfInt.
But we want such a method to be bound the instance x
as well. So a generic method in ListOf needs TWO object
arguments:

	1) the first argument binds to ListOfInt
	2) the second argument binds to x

Lets consider an 'append' method:

	class ListOf:
		# ...
		def append(erep, obj, elem):
			if erep.etype is not type(elem):
				raise "Attempt to add element of wrong type to List"
			else:
				obj.data.append(elem)

As you can see, we have not only added a user defined 'type'
to the type system by a new technique (other than using a class instance),
but we have added a whole family of them. We have added a new kind
of type: a generic type. 

	WE HAVE DONE IT ENTIRELY IN (V)PYTHON.

Why am I shouting? Why is this important? Obviously, it is
useful to have control of Python IN Python: Python is not
only easier to write than C (or Java), it is also more portable.

But there is another reason. Viper is intended to be compiled.
Defining the type system in 'native code', whether it be C,
Java, or Ocaml, or whatever, makes compilation much harder:
all the details of the native code interpreter implementation
have to be preserved in the generated code. It is difficult
to optimse the code, because it is likely to be complex.

But if the type system is written in Python, then the compiler
doesn't need any special work to compile the type system,
it can concentrate on optimising the core language,
and treat the type system as a pure python library component.
Consequently, user defined 'native types' can be quite efficient,
because they're compiled.

In fact, at the moment, user defined types are MORE efficient
than builtin ones! The reason is that the builtin types cannot
have a pointer to their type objects when they're constructed,
because these type objects don't exist when the interpreter is
started: they're created when the interpreter imports the module
'py_types' at startup, and while that importation is hard coded
into the interpreter, the file is parsed and executed like any
other python module: indeed, the same applies to exceptions.

Consequently, builtin types have a __typename__ attribute which
names the type as a string, and the type is obtained by
looking up in the py_types module, using that name as the
dictionary key. This extra indirection is not required for
client type instances, for which the 'new_object' method
can set the type object directly.

Just to re-emphaises: I have NOT tried the generic typing
scheme yet. However, the 'types are arbitrary objects' is already
implemented and works: the types of all my builtin types
are defined in python in py_types.py, the python
module 'types' executes successfully,and some cursory
assertions in a test file yield the expected behaviour,
including the ability to call type methods as object methods.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From hschroeder at gmx.net  Sat Aug  7 15:36:04 1999
From: hschroeder at gmx.net (Henning Schroeder)
Date: Sat, 7 Aug 1999 21:36:04 +0200
Subject: Context-sensitive help for IDLE.
Message-ID: <4r1io7.a43.ln@odin.henning-schroeder.de>

Hi there!

I added context-sensitive help to IDLE. Don't expect perfect code but it works.
You just have to press F2 in the editor window und Netscape Navigator shows the
documentation for the word under your cursor. Besides the help-menu has two new
entries: Language Reference and Python Tutorial.

If you are interested download the extension from
  http://homepages.emsnet.de/~schroed/python/idle-0.5.1.tgz
The archive also contains the latest IDLE version from David Asher's daily
cvs-snapshots. The following files belong to the extension:
  ContextHelp.py
  helpindex.px
  idlerc.py
  extend.py 
You can also use these files with an older version of IDLE.

Maybe you have to edit idlerc.py in order to change the doc-path. The first 
time you press F2, an index file (pyhelp.idx) of the keywords will be created -
so keep patient.


Greets
  Henning

---
ich at henning-schroeder.de  *  http://www.henning-schroeder.de



From molivier at my-deja.com  Tue Aug 31 14:39:15 1999
From: molivier at my-deja.com (molivier at my-deja.com)
Date: Tue, 31 Aug 1999 18:39:15 GMT
Subject: SMTP server source code
Message-ID: <7qh7gh$1ru$1@nnrp1.deja.com>

Looking for python source for a simple SMTP server implementation.
Ideally would use threads.  Any pointers/sources appreciated.

--Michael


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From chad at vision.arc.nasa.gov  Wed Aug 11 21:01:21 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Thu, 12 Aug 1999 01:01:21 +0000
Subject: Problem when using FileDialog from PMW Dialog.  Quick fixes?
Message-ID: <37B21CE1.488F45FA@vision.arc.nasa.gov>

I have an application where I open a Python Mega Widget dialog (local,
not global) and within it I later open a FileDialog() for file selection
(Don't ask, why, but calling a nested dialog makes sense for the app.)
In any case, it doesn't work.  When the FileDialog exits, the Pmw Dialog
cannot get focus back (even with some code to try to force it.)

Before I pursue this further, can anyone tell me if it is even worth
trying?  It is nice to use the canned FileDialog, but its implementation
is somewhat crufty (IMHO) and I can probably work around it in other
ways (Creating my own FileDialog from PMW widgets is a nice possibility.
)   But if there is a quick fix, I'd like to hear it.

Below you'll find  some psuedo-code to show roughly what I'm doing.  Run
the code, click on the "List" button to activate the FileDialog, then
exit the FileDialog.  Hitting the PMW Dialog "Cancel" button doesn't
work (as it does if you never open the FileDialog).  Pardon the rough
code :)  You will probably have to kill the process if you trigger
the problem, so be warned.

Chad Netzer
chad at vision.arc.nasa.gov

------------------------------ [ Cut Here ]
------------------------------
from Tkinter import *
import Pmw
from FileDialog import LoadFileDialog

def CreatePMWDialog(parent):
    global dialog
    dialog = Pmw.Dialog(parent,
                        buttons = ('OK', 'Cancel'),
                        defaultbutton = 'OK',
                        title = 'A test: The FileDialog will steal focus
forever',
                        command = DialogAction)
    dialog.parent = parent

    interior = dialog.interior()
    entry_frame = Frame(interior)
    list_but = Button(entry_frame, text="List", command=ListFiles)

    list_but.grid(row=0, column=1, sticky=N+S)
    entry_frame.pack(side=TOP, anchor=NW)

    # Now execute the dialog
    dialog.activate()
    return

def ListFiles():
    global dialog
    dialog = LoadFileDialog(dialog.parent)
    file = dialog.go("")
    print file

def DialogAction(button_name):
    global dialog
    if button_name == 'Cancel':
        dialog.deactivate()
    elif button_name == 'Ok':
        dialog.deactivate()
    return

# Test it by parsing a prm file
if __name__ == '__main__':
    tk_root = Tk()
    foo = CreatePMWDialog(tk_root)
    tk_root.mainloop()





From tim_one at email.msn.com  Sat Aug 28 04:05:31 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 28 Aug 1999 04:05:31 -0400
Subject: Returning none
In-Reply-To: <slrn7scf7d.pa.clgonsal@keeshah.penguinpowered.com>
Message-ID: <000101bef12c$19566520$ad2d153f@tim>

[C.Laurence Gonsalves]
> def seqToString( seq ):
>     ...
>     result = result + ']'
>
> ...
> Personally, I think falling out of functions and expecting None to be
> implicitly returned is a very Perl-ish thing to do...

Oddly enough not.  In the absence of an explicit "return", Perl returns the
value of the last expression evaluated before falling off the end, so in your
function above it returns what you intended.

> Why don't we add $_ to Python as well?

Because that would be Perlish <wink>.  The notion of an implicit operand first
appeared in string matching languages, where it's a PITA (and error-prone) to
have to keep specifying the same string and index vrbls.  Perl buys efficiency
& convenience with this trick for the same reason, although outside of string
crunching I'm not sure I've seen a good use of it.  The reason it doesn't fit
in Python is that Python isn't specialized to string crunching -- or to any
other particular domain.

> I don't think it's too hard to explicitly return None when that's
> what you want to do, and I would think/hope that most existing Python
> code does this already.

Python doesn't return None for the convenience of people too lazy to explicitly
return None themselves; it's instead a cheap way to make it probable that
someone using a procedure as a function will get an error (None supports no
operations except true/false interpretation) without blowing up the interpreter
(do this in C and you'll access random stack trash).

> It is possible, but more complicated, to make this a compile time
> error.  This would essentially require a function/procedure
> distinction with restrictions about how you used return in each one.
> ...

Do you imagine Guido didn't consider this when he designed the language?

nothing's-a-pure-win-ly y'rs  - tim





From timcollinsdeja at my-deja.com  Fri Aug 13 14:14:54 1999
From: timcollinsdeja at my-deja.com (timcollinsdeja at my-deja.com)
Date: Fri, 13 Aug 1999 18:14:54 GMT
Subject: MySQL module crashes
References: <37B2F23A.7E084E5C@icrf.icnet.uk>
Message-ID: <7p1na7$8qu$1@nnrp1.deja.com>

Actually, I had a similar problem when trying to use MySQL with a Java
App Server. Apparently, The JDBC driver is completely solid. The other
drivers seem to be of varying degrees of usefulness...


In article <37B2F23A.7E084E5C at icrf.icnet.uk>,
  Arne Mueller <a.mueller at icrf.icnet.uk> wrote:
> Hi All,
>
> I've just installed MySQL 1.4, mysql 3.22.25 and python 1.5.2b2
>
> Unfortunately python coredumps :-(
>
> mueller at jumbo:~/src/Python-1.5.2b2> python
> Python 1.5.2b2 (#17, Aug 12 1999, 16:53:47) [C] on irix6
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> python> import MySQL
> Fatal Python error: PyThreadState_Get: no current thread
> Abort (core dumped)
>
> I'm runniing irix 6.5 on R5000/10000. Does anybody have any
> experience with mysql module in the above version combination?
> In general I've no problems with threads in python.
>
> 	thanks for help,
>
> 	Arne
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From varner at PENELOPE.CHEM.UIC.EDU  Mon Aug  2 15:53:30 1999
From: varner at PENELOPE.CHEM.UIC.EDU (varner at PENELOPE.CHEM.UIC.EDU)
Date: 02 Aug 1999 14:53:30 -0500
Subject: Interactive use of Tk
References: <3798A5D7.674D@cs.bham.ac.uk> <3798D48D.CE9BA6A3@raima.com>
Message-ID: <82k8reot6t.fsf@PENELOPE.CHEM.UIC.EDU>

Thanks Alexander.  I've been suffering from this problem for a while.
But now I have a similar problem with Pmw.  Here's a tiny class that
displays the problem (and not much else).

### 
from Tkinter import *
import Pmw

class no_graph (Tk):
    def __init__ (self):
        Tk.__init__ (self)
        Pmw.initialise (self)
        self.graph = Pmw.Blt.Graph (self, title = "No Data")
        self.graph.pack ()

        button = Button (self, text = "Quit", command = self.destroy)
        button.pack ()

    def destroy (self):
        self.withdraw ()
        Tk.destroy (self)
        self.quit ()
###

Interactively, the second time I call no_graph (), I get:

Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "no_data.py", line 8, in __init__
    self.graph = Pmw.Blt.Graph (self, title = "No Data")
  File "/usr/local/lib/python1.5/site-packages/Pmw/Pmw_0_8/lib/PmwBlt.py", line 215, in __init__
    Tkinter.Widget.__init__(self, master, _graphCommand, cnf, kw)
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 1084, in __init__
    self.tk.call(
TclError: invalid command name "::blt::graph"

I'm guessing I have to quit Pmw before I reinitialize, but I don't
know how.  Does anyone know of a good online reference for Pmw or Blt?
Thanks.

-- 
Sam Varner
SEL 2210 (MC 111)
Department of Chemistry
University of Illinois at Chicago



From robin at jessikat.demon.co.uk  Thu Aug  5 21:31:34 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Fri, 6 Aug 1999 02:31:34 +0100
Subject: Linux why not fully dynamic?
Message-ID: <aOAnwBA2rjq3EwnZ@jessikat.demon.co.uk>

Why doesn't ./configure set up for a libpython.so.1.5.2 or whatever?
Also does anyone understand how the RH setup knows where the .py & .so
files will be stored?
-- 
Robin Becker



From tchrist at mox.perl.com  Thu Aug 19 11:22:49 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 08:22:49 -0700
Subject: More random python observations from a perl programmer
Message-ID: <37bc1339@cs.colorado.edu>

Per your collective requests, here is one of the documents
with my collective observations from learning Python.

This article was posted to the Perl newsgroup, but I have duplicated
the posting here without crossposting to avoid flame wars.

--tom

SIMILARITY:
    When Python talks about tuples, lists, and dictionaries, 
    Perl people should think of lists, arrays, and hashes
    respectively.  Except that tuples in python are 1st-class
    citizens.  (Unclear whether this is good; see below.)

COOLNESS:
    Python's print() function is more like the Perl debugger's
    "x" command; that is, it's recursive and pretty-printed.

GOTCHA: (medium)
    The whole mutability versus immutability thing causes many 
    confusions.  For example:
	t = ([1,2],3)
	t[0][0] = "fred"	# legal
	t[1]    = "fred"	# ILLEGAL
    I don't understand why tuples weren't just lists that weren't
    somehow marked "read-only" or "constant".   That way you
    could do the same with dictionaries or any other data types.

GOTCHA: (low)
    You can't use "in" on dicts.  Instead, you must use 
	d = { "fred":"wilma", "barney":"betty" }
	if d.has_key("fred"):	# legal
	if "fred" in d:	        # ILLEGAL
    I don't understand why this was done.

SIMILARITY:
    Both "and" and "or" return their last evaluated value
    in both Python and Perl.

GOTCHA: (low)
    I don't understand why we have so many C operators, such as "&" and
    "|" and "~", but we don't have "&&" or "||" or "!"  We also have "<<"
    and ">>".  It's bizarre that the relational and "?:" are missing.
    Python uses so much from C to make people see what they expect to see,
    but this is missing.  Why?

GOTCHA: (high)
    Local variables are never declared.  They just *happen*.  If a
    variable is assigned to, then it belongs to that "scope".  If it's
    only looked at, it may be from the current global scope.

GOTCHA: (medium)
    If you have a function that expects two arguments, such as:
	def fn(x,y):
	    return x + y
    and you have a tuple of the two elements:
	t = (1,2)
    You can't just call that function:
	print fn(t)
    Or you get a "TypeError: not enough arguments; expected 2, got 1"
    error.  You have to use this apply() function to get around this
    issue:
	print apply(fn,t)
	3

SIMILARITY:
    The number zero, the empty string, and the special value None
    are all false.  Non-zero numbers and non-empty strings are
    all true.  This is all like Perl.  But here an empty tuple () 
    is false, as is an empty list [] or an empty dictionary {}.
    In Perl, the last two are true if you write them that way, because
    they're references.  

GOTCHA: (medium) 
    Assignment is a statement, not an operator.  That means that
    people are constantly writing loops that aren't really testing
    the exit condition, as in:
	while 1:
	    line = os.readline()
	    if not line: break
	    .....
    Because they can't write:
	while (line = os.readline()):	# ILLEGAL
    I feel that this hides the real test condition in a place
    it doesn't belong.

GOTCHA: (low) 
    There are no compound assignment operators, like +=.

DISSIMILARITY: 
    There is no "test at the bottom" loop as in C's do{}while.

DISSIMILARITY: 
    There is no "three part for" loop from C.

GOTCHA: (low)
    There are no loop labels, and therefore "break" and "continue" are
    only through the next level.  This encourages the proliferation of
    spurious boolean condition variables.  It was annoying when Pascal
    made you do the same thing.  There is no "goto", which is how C
    works around it.  As with many things in Python, here you force the
    user to be tangled up with exceptions just to do very simple things.
    That's not as clear a solution.

SIMILARITY:
    Both Python and Perl are 0-based for indexing.

DISSIMILARITY:
    Python uses [] to create a new empty list, which you
	then reference with [] subscripts.
    Perl uses [] to create a new empty array, which you
	then reference with [] subscripts.
    Python uses {} to create a new empty dictionary, which you
	then reference with [] subscripts.
    Perl uses {} to create a new empty hash, which you
	then reference with {} subscripts.

GOTCHA: (high)
    I had hoped that having *only* reference semantics would spare the beginner
    from having to keep in mind the difference between a reference and its
    referent as we have in C or Perl.  Alas, it is not to be.
	 x = [1,2,3]
	 y = x
	 x[1] = "fred"
	 print y
	[1,"fred",3]

COOLNESS:
    You can "foreach" across multiple items.
	a = [ [1,2], [3,4], [5,6] ]
	for (i,j) in a:
	    print i+j
      3
      7
      11

COOLNESS:
    Python's "longs" are actually built-in, arbitrary precision 
    integers (i.e. BigInts)

COOLNESS: (?)
    Named parameters are built into the language.  So are 
    default paramters and variadic ones, although order matters
    of course.  It's quite elaborate.

COOLNESS:
    Namespaces seem to be *the* thing in this language.  You can always
    get a listing of anything's names/attributes, even all the built-ins
    (which you can't do in perl) or all the names in the local scope
    (which you also can't do in perl).  

DISSIMILARITY:
    A class method call gets no class name as its implicit extra
    argument the way an instance method would.  

DISSIMILARITY:
    In Python, there is no difference between single and double quotes:
    both interpolate C-style escapes like \t, and neither 
    interpolates variables.  In Perl, single quotes do neither escapes
    nor variables, but double quotes do both.

GOTCHA: (low)
    Single and double quoted strings can't cross line boundaries.
    You need triple quotes for that!

SIMILARITY:
    Python does a depth-first recursive search on class ancestors
    looking for resolution of a method call.  So does Perl.

DISSIMILARITY:
    Python also does an inheritance search for data members that
    are missing.  This is probably cool, however.  There is no
    big difference between a functional and a data member.

DISSIMILARITY:
    There are no private variables in a module or a class.
    They can always be fully-qualified and accessed from
    without.  Perl can do this with file-scoped lexicals,
    but still can't do so with data members (modulo obscure
    tricks such as using a non-hash for an object).

GOTCHA: (medium)
    Perl's hex() function is the opposite of Python's.  Perl's hex
    function converts 22 into 37.  Python's hex function converts 37 into
    0x22.  Oh my!  Like hex, Python has oct() completely backwards from
    Perl.  Python's oct(44) returns 054, but Perl's oct(44) return 36.

SIMILARITY:
    Constructors and destructors (__init__ and __del__) aren't 
    automagically called in parent classes in Python.  Neither
    in Perl.

DISSIMILARITY:
    Because there are no variable markers in Python, there are no 
    variable interpolation in strings.  That makes something like
	print "this $x is $y here\n" 
    become in python
	print "this %d is %d here\n" % (x, y)
    But since "stringification" is handy, Python usurps backticks
    for that purpose:
	print "this is", `x`, "here"
    Is going to call the x object's stringification method (x.__str__)
    to get a print value.  In perl:
	print "this is $x here"
    also does so, although the stringification method is oddly named
    in Perl.
	use overload '""' => ....

GOTCHA: (medium)
    Things that return lists can't be used where a tuple is expected.
    A function that returns a list must be coerced into a tuple to 
    use this, though.
	def fn: return [1,2]
	print "this %d is %d here\n" % fn()  # ILLEGAL
	print "this %d is %d here\n" % tuple(fn())
    The illegal part points out that this is an 
	TypeError: illegal argument type for built-in operation
    Which isn't very helpful.  

GOTCHA: (high)
    Python has no manpages!  The horror!!!!!!!!!!!!!!!!!!!!!!
    ENODOC

GOTCHA: (low)
    Often Python's error messages leave something to be desired.
    I don't know whether 

GOTCHA: (medium)
    All ranges are up to but *not including* that point.  So range(3)
    is the list [0,1,2].  This is also true in slices, which is the
    real gotcha part.  A slide t[2:5] does not include t[5] in it.

SIMILARITY:
    Negative subscripts count back from the right.  Same as in Perl.

COOLNESS:
    Slices can omit either the starting point or the ending point.
	t[2:5]		# elts 2 through 4 (but not 5!)
	t[:5]		# elts up to 4 (but not 5!)
	t[:-1]		# all but the last element (up to but not to -1)

DISSIMILARITY:
    Python doesn't convert between strings and numbers unless you tell
    it to.  And if you don't tell it to, you get an exception.  Therefore
    you have no idea what these 
	x = y + z
	x = y - z
	x = y * z
    are really doing.  The first would concat strings or lists, the last would
    repeat them.  Personally, I don't care for this, because I always
    wonder what subtracting one string or list from another would be.

COOLNESS: 
    Python has a reduce() built-in that works somewhat like map().

COOLNESS: 
    Python's filter() [same-ish as Perl's grep()] and its map() operators
    can operate on items from separate sequences in parallel.

GOTCHA: (high)
    This is a big surprise: strings are not atomic (although they are
    immutable).  They are instead sequences of characters.  This comes
    up in strange places.  For example:	
	>>> for c in ("weird", "bits"):
	...      print c
	... 
	weird
	bits
	>>> for c in ("weird"):
	...      print c
	... 
	w
	e
	i
	r
	d
    The second case autosplit the characters!  Here's another:
	>>> print map(None, "stuff")
	['s', 't', 'u', 'f', 'f']
	>>> print map(None, "stuff", "here")
	[('s', 'h'), ('t', 'e'), ('u', 'r'), ('f', 'e'), ('f', None)]
    (Python's map None is like Perl's map $_.) 

DISSIMILARITY:
    This also means that instead of substr() in Perl, you slice a string
    as in 
	s = "string"
	print s[2:4]
      ri
    Yes, that's all you got.  Strange, eh?  See below on ranges.


GOTCHA: (high)
    Because everything is a reference, and there's no way to dereference
    that reference, it turns out that there is no trivial way to copy
    a list!  This fails:
	x = [1,2]
	y = x
    Because you don't get a new list there, just a copy to an
    old one.  Suggested work-arounds include
	y = x[0:]
	y = x[:]
	y = x + []
	y = x * 1
    This forces people to think about references, again.    
    So much for lists being first class citizens!  Compare 
    this with Perl's
	@x = (1,2);
	@y = @x;
    Or even with references:
	$x = [1,2];
	$y = [ @$x ]; 
    or 
	@y = @$x;

GOTCHA: (medium)
    Slices in Python must be contiguous ranges of a sequence.
    In Perl, there's no such restriction.  

GOTCHA: (medium)
    You can't slice dictionaries at all in Python.  In Perl, it's
    easy to slice a hash, and just as sensible.

GOTCHA: (high)
    As we saw with lists, because everything is a reference, and there's
    no way to dereference that reference, this means that again there
    is also no built-in, intuitive way to copy a dictionary.  Instead, 
    the suggested work-around is to write a loop:
	new = {}
	for key in old.keys:
	    new[key] = old[key]
    But this is guaranteed slower, because it's not at the C level.
    It also shows that dictionaries aren't first class citizens in python
    as they are in Perl:
	%old = ( "fred" => "wilma", "barney" => "betty" );
	%new = %old;
    or even with references:
	$old = { "fred" => "wilma", "barney" => "betty" };
	$new = { %$old };
    or 
	%new = %$old;

GOTCHA: (high)
    Lists don't autoallocate the way dictionaries do.  So while this works
    on two dictionaries:
	new = {}
	for key in old.keys:
	    new[key] = old[key]
    This fails on two lists:
	new = []
	for i in old:
	    new[i] = old[i]
    Because Python refuses to grow the list as it did the dictionary.  Perl
    will grow both arrays and hashes just fine.

GOTCHA: (medium)
    There's no way to set up a permitted exports list.  The caller may have
    anything they ask for.  

COOLNESS:
    DBM files seem (?) to automatically know about nested datatypes.   

GOTCHA: (medium)
    Importing a variable only gives you a local copy.  In Perl, it makes
    two names for the same object.

GOTCHA: (low)
    Because you can't declare a local variable, you can't create variables
    local to something other than a module (ie. file), class, or function.
    There are no loop-local variables; they would persist through the end
    of the function.  You can't have a scope you can't name (well, almost;
    see lambdas)

GOTCHA: (high)
    Scopes don't nest in Python, but they they do in Pascal, Perl, or C.
    This is supposedly "simpler", but it's very suprising in many ways:
	x = 1			# global
	def fn1(): 
	    x = 10		# implicit local
	    def fn2():
		print x		# whose?
	    fn2()
	fn1()
    The answer is that it prints 1, because fn2 has no x in its local
    scope, so gets the global.  The suggested work-around is
	    def fn2(x=x):
    using default parameters.

GOTCHA: (medium)
    You can't cut and paste these examples because of the issue
    of white space significance. :-(

GOTCHA: (low)
    List objects have built-in methods, like 
	l.append(x)
    But string objects don't.  You have to import
    from the string module to get them, and then they're
    functions, not methods.

GOTCHA: (low)
    You have insert and append methods for lists, but only
    a del function.  Why is it a function not a method, and
    why isn't it spelled out?  Apprently the fashionable wayto do this is
    now
	a[2:3] = []
    Which of course, deletes only one element.  ARG.

DISSIMILARITY:
    Where Python says "elif", Perl says "elsif".

SIMILARITY:
    Neither Perl nor Python supports a case or switch statement, requiring
    multiway if's or a lookup of a function dispatch table.

DISSIMILARITY:
    Where Python says "break" and "continue", Perl says "last" and "next".

DISSIMILARITY:
    Where Perl says "continue", Python uses "else" (for a loop block).

DISSIMILARITY:
    Constrast Python:
	import os
	data = os.popen('grep %s %s' % (patstr, srcfile)).read()
    with Perl:
	$data = `grep $patstr $srcfile`;

GOTCHA: (high)
    There doesn't seem to be away other than using low-level hand-rolling
    of posix functions to supply things like os.popen and os.system
    a list of shell-proof arguments.  Looks like it always goes through the
    shell.  This has security ramifications.  

GOTCHA: (high)
    Because you can't use readline() to get a number, people seem to enjoy
    calling eval() just to get a string turned into a number:
	import sys
	str = sys.stdin.readline()
	num = eval(x)
    This is scary.  Even scarier is the propensity for calling input(),
    which auto-eval()s its input to make sure it's the right "type".
    (Funny that a soi-disant "typeless language" should be so danged
    picky about this.)  That means you see people write:
	num = input("Pick a number? ")
    But the person can supply as an answer 0x23 or 2+9 or pow(2,4)
    or os.system("rm -rf *").  I am stunned.

GOTCHA: (medium)
    The expression 3/4 is 0, which is false.  In Perl, 3/4 is 0.75,
    which is what you'd expect.  You need to force the floatness.
    Sometimes Python is just too tied to C, other time not enough.
    This is one of the former.

GOTCHA: (low)
    Regexes default to emacs style, not egrep style!  Gads!  This is
    surprising to anyone except an emacs weenie.  Sigh.  And just *try*
    to read the manpage on the differences based on passed in arguments
    establishing the style.  There aren't any manpages!  Have a nice day.

GOTCHA: (low)
    An out-of-bounds list reference raises an "IndexError: list index out
    of range" exception, but not if you use a slice to get at it!
	t = range(5)
	print t[2:17]
      [2, 3, 4]

COOLNESS:
    Relationals stack:
	x < y < z
    means 
	x < y and y < z

COOLNESS:
    There's no distinction between the mechanism used for operator
    overloading and tying, as there is in Perl.  Both use special
    method names.  

GOTCHA: (high)
    Python's lambda's aren't really lambdas, because they are only 
    expressions, not full functions, and because they cannot see
    their enclosing scope.  

DISSIMILARITY:
    What Perl calls eval(), Python calls exec().  
    What Perl calls exec(), Python calls os.execv().  

GOTCHA: (low)
    Python's eval() only works on expressions, not full code blocks full
    of statements.  You need exec() for the whole thing.

GOTCHA: (medium)
    This is a tuple of two elements
	(1,2)
    But this is a tuple of one element:
	(1,)
    Whereas this is merely the expression 1:
	(1)
    Yes, the trailing comma counts.

GOTCHA: (low)
    Normally, a print statement in python adds a newline.  If you
    don't want one, you need a trailing comma!

DISSIMILARITY:
    Python uses a pow() function for Perl's ** operator.
    Wait, no, they just added "**" later, with the same
    semantics as in Perl, except that something like 2**3**4
    in Perl gives you an answer, and in python, an exception.

GOTCHA: (low)
    Python has a round() function, but it seems to ignore the 
    IEEE rules of rounding.  However, its sprintf operator doesn't:
	>>> print "int of %f is %.0f" % (1.5, 1.5) 
	int of 1.500000 is 2
	>>> print round(1.5,0)
	2.0
	>>> print "int of %f is %.0f" % (2.5, 2.5)
	int of 2.500000 is 2
	>>> print round(2.5,0)
	3.0
    And I'd jolly well like to know why I wasn't allowed to use
	print "int of %f is %.0f" % (2.5) * 2
    or if needed, 
	print "int of %f is %.0f" % ( (2.5) * 2 )

GOTCHA: (medium)
    Anything that python doesn't like, it raises an exception about.
    There is no fail-soft.  Even non-exception issues raise exceptions.
    It's pervasive.  K&P curse languages that force people who want
    to open() a file to wrap everything in exception code, saying that
    "failing to open a file is hardly exceptional".  

GOTCHA: (medium)
    You can't just interchange tuples and lists in python the way
    you can lists and arrays in Perl.  This is an error:
	import sys  # otherwise, no argv for you, buddy
	print "First arg %s and second %s" % sys.argv[1:3]
    because you need
	print "First arg %s and second %s" % tuple(sys.argv[1:3])

GOTCHA: (low)
    I can't figure out how to write a class destructor for at exit
    handling the way I can with Perl's END{}

GOTCHA: (medium)
    Python has no 2nd GC pass at thread shutdown time to find lost
    objects and destruct them.  This is a problem on embedded systems,
    and breaks correctness issues.

SIMILARITY:
    Like Perl, Python requires access to members through a self reference.
    This gets rid of C++'s icky hidden scope.  This is good.  

COOLNESS:
    I believe that Python checks the prototype signature on methods
    as well as on functions.  I wonder whether Tim Bunce's ideas run-time
    evaluation of prototypes for Perl might be able to do this.

GOTCHA: (low)
    sort and reverse are in-place.  This leads to verbose code, such as:
	old = [1,2,3]
	new = old
	new.reverse()
    Likewise for sort.  

GOTCHA: (low)
    You have to compiled the definition for a  function before you may
    compile the call it.  This is ok
	def fn(): 
	    print "hi"
	fn()
    But putting the fn() call first:
	fn()
	def fn(): 
	    print "hi"
    Produces the ever helpful: 
	NameError: fn
    error message.  So much for autoloading.  Then again, this would
    save me from the darned legnth() bug in Perl, that is, switching
    characters.  Maybe a "no autoload" pragma for perl?  I always though
    "use strict subs" shoud ldo this.  But another python mystery is
    why it's ok to compile code to methods that you don't know whether
    will be legal, but not functions.  
	ob = some_fn()
	ob.scattter()
    That will compile, even though I used an extra 't'.  It will die
    on run-time, just as in Perl.  But the fn() case at the start 
    of this looked like a compile-time death.
    ADDENDA: It turns out that def() happens *AT RUN TIME* in python.
    That's the root of this problem.  In Perl, a sub{} is a compile-time
    thing.

GOTCHA: (low)
    When you need to make an empty copy of the same type, you write
	x = y[:0]
    So much for being typeless. Sigh.

SIMILARITY:
    In theory, the Python rexec() module should be like Perl's Safe
    module.  I haven't tested to see whether they chroot the namespace,
    or whether the evil method name leakages occur through the box.

COOLNESS:
    Any function of class can have a significant string called a "doc
    string" that can be accessed as whatever.__doc__

QUESTION:
    What is and what is not thread safe?

SIMILARITY:
    Python uses None in many places where Perl uses undef().
    But not all.  You can't over-subscript an array and get
    away with it.

-- 
    I know it's weird, but it does make it easier to write poetry in perl.    :-)
                    --Larry Wall in <7865 at jpl-devvax.JPL.NASA.GOV>



From mal at lemburg.com  Thu Aug 19 17:02:35 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Thu, 19 Aug 1999 23:02:35 +0200
Subject: ANN: mxProxy Package - Version 0.2.0
Message-ID: <37BC70EB.7E03BBEB@lemburg.com>

ANNOUNCING:

                         mxProxy Version 0.2.0

           A Python extension package providing a configurable
              transparent proxy object implementation with
              fine grained access control, data hiding and
                  weak references for Python objects

WHAT IT IS:

mxProxy uses an extension type to wrap objects that you pass
to its contructor in essentially two different ways:

1. via a strong reference which works just like any other
   reference in Python or

2. using a weak reference which allows you to build circular
   reference models in your data without producing the other-
   wise difficult to handle memory leakage induced by this.

Furthermore, the implementation offers some very convenient
access and cleanup protocols which help you control access to
the object (even on very low levels such as type slots) and
provide automatic mechanisms for object cleanup prior to
normal garbage collection.

Wrapping is done in a transparent way, so that you normally
don't even have to recode programs to have them use the
proxy instead of the real object -- only the type possibly
changes.

mxProxy enables you to build secure environment without
having to use techniques like the standard library module
Bastion or restricted environments on a per-object basis.

Several different higher level interfaces for the generic type
are included in the package, such as InstanceProxy, caching
mix-ins, etc. 

WHAT'S NEW ?

The 0.2.0 release added a powerful weak reference mechanism
to the package and also fixed some minor bugs.

This release also includes a precompiled Windows DLL so that you can
use the package right away. Installing on that platform boils down
to a simple unzip in the \Python\Lib directory.

WHERE CAN I GET IT ?

The full documentation and instructions for downloading and installing
can be found at:

    http://starship.skyport.net/~lemburg/mxProxy.html

WHAT DOES IT COST ?

It comes with a Python-type license, but is free otherwise.

WHERE CAN I GET SUPPORT ?

I am offering commercial support for this package through
Python Professional Services Inc. (http://www.pythonpros.com).
Look on their support pages for details or contact me directly.

REFERENCE:

<P><A HREF="http://starship.skyport.net/~lemburg/mxProxy.html>
mxProxy 0.2.0</A> - generic proxy object providing low-level
access control and weak references. (19-Aug-1999)

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   134 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From watterss at pilot.msu.edu  Sun Aug 22 15:10:24 1999
From: watterss at pilot.msu.edu (Shannon Watters)
Date: Sun, 22 Aug 1999 15:10:24 -0400
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com> <37bfe42c$0$221@nntp1.ba.best.com>
Message-ID: <7ppi33$8oe$1@msunews.cl.msu.edu>

Xah <xah at best.com> wrote in message news:37bfe42c$0$221 at nntp1.ba.best.com...
>
> Perl wouldn't be a nice community without bad asses like Tom Christensan,
> Chris Nador and some other regulars I don't want to mention now. We thrive
> on smiting newbies. We live to smite each other too. If you start to feel
> uncomfy,  get out.
>
> un'stand?
>

It's easy to act like a tough guy over the net, isn't it?





From Delius at ebox.de  Fri Aug  6 08:36:26 1999
From: Delius at ebox.de (Felix von Delius)
Date: Fri, 6 Aug 1999 14:36:26 +0200 
Subject: AW: Help with dictionary, please
Message-ID: <7B685231630CD311A9B20090276CE41F0B901E@eboxserver2.harc.de>

Hi Rob,

I tried the example and it worked for me. Maybe it's just a problem of
indentication, because the last two lines in Your example should be indentet
at the same level and you indented two more spaces in front of
'dict[key]...' ?!?

> 	if prev1 != '' and prev2 != '':
> 	 	   key = (prev2,prev1)
> 	           dict[key] = dict[key] + 1

Or You're mixing spaces and tabs?!? Just a guess...

-Felix
---
Dipl.-Ing. Felix von Delius		Fon: 0911/4244-110
Projektleiter-IT				Fax: 0911/4244-100
EBox GmbH - Agentur f?r neue Medien	Mobil: 0172/8261220
Vordere Cramergasse 11		Email: delius at ebox.de
D-90425 Nuernberg			Web: http://www.ebox.de




From hschroeder at gmx.net  Wed Aug 11 07:09:14 1999
From: hschroeder at gmx.net (Henning Schroeder)
Date: Wed, 11 Aug 1999 13:09:14 +0200
Subject: Resuming with (ftplib/httplib)?
References: <37AF69CC.3ECFC8EE@sprint.ca>
Message-ID: <qklro7.us.ln@odin.henning-schroeder.de>

Orlando Vazquez <ovazquez at sprint.ca> wrote:
> So my question is, how can I, using ftplib and httplib, resume a
> download? Is it possible? Has this been done before? Any code someone
> would be willing to share? I would appreciate it soo much. :-)
HTTP version 1.1 supports byte ranges. There is an interesting article in
ApacheWeek (http://www.apacheweek.org) about HTTP 1.1.
The ftp-protocoll has a REST command to resume transfers. See the hack 
below which works for me.

  Henning

------------------------------------------------------------
# Module xftplib.py
from ftplib import *

class XFTP(FTP):

    def size(self, filename):
	self.voidcmd("TYPE I") #seems necessary to get the *right* size	
	return FTP.size(self, filename)
		
    def rest(self, start):
	resp = self.sendcmd("REST %s" % start)
	if resp[:3] <> "350": raise error_reply, resp

    def ntransfercmd(self, cmd, resume):
	'''Initiate a transfer over the data connection.
	If the transfer is active, send a port command and
	the transfer command, and accept the connection.
	If the server is passive, send a pasv command, connect
	to it, and start the transfer command.
	Either way, return the socket for the connection and
	the expected size of the transfer.  The expected size
	may be None if it could not be determined.'''
	size = None
	if self.passiveserver:
	    host, port = parse227(self.sendcmd('PASV'))
	    conn = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
	    conn.connect(host, port)
	    if resume: self.rest(resume)
	    resp = self.sendcmd(cmd)
	    if resp[0] <> '1':raise error_reply, resp
	else:
	    sock = self.makeport()
	    if resume: self.rest(resume)
	    resp = self.sendcmd(cmd)
	    if resp[0] <> '1':
		raise error_reply, resp
	    conn, sockaddr = sock.accept()
	    if resp[:3] == '150':
		# this is conditional in case we received a 125
		size = parse150(resp)
	    return conn, size

    def transfercmd(self, cmd, resume=0):
	'''Initiate a transfer over the data connection.  Returns
	the socket for the connection.  See also ntransfercmd().'''
	return self.ntransfercmd(cmd, resume)[0]

    def retrbinary(self, cmd, callback, resume=0, blocksize=8192):
	    '''Retrieve data in binary mode.
	    The argument is a RETR command.
	    The callback function is called for each block.
	    This creates a new port for you'''
	    self.voidcmd('TYPE I')
	    conn = self.transfercmd(cmd, resume)
	    while 1:
		    data = conn.recv(blocksize)
		    if not data:
			    break
		    callback(data)
	    conn.close()
	    return self.voidresp()



From mwh21 at cam.ac.uk  Tue Aug 31 13:08:35 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 31 Aug 1999 18:08:35 +0100
Subject: Soundex implementation (was: RE: Fuzzy string matching?)
References: <000001bef12c$139c0680$ad2d153f@tim> <19990831123108417.AAA352.267@max41072.izone.net.au>
Message-ID: <m3btbn98to.fsf@atrus.jesus.cam.ac.uk>

"John Machin" <sjmachin at lexicon.net> writes:

> <FontFamily><param>Courier New</param>O<bigger>

What's that? Some bastardized form of html? rtf? 

Can you try again in plain text please?

TIA,
Michael




From peter.stoehr at weihenstephan.org  Thu Aug 19 15:02:22 1999
From: peter.stoehr at weihenstephan.org (Dr. Peter Stoehr)
Date: Thu, 19 Aug 1999 21:02:22 +0200
Subject: Canvases not resizable?
References: <37BC2305.B3C@cs.bham.ac.uk>
Message-ID: <37BC54BE.FBA34180@weihenstephan.org>

Hi Stuart,

Stuart Reynolds wrote:

> Are Tkinter canvases fixed to one size? I don't seem to be able to get
> them to expand to fill their container when its resized even the though
> the canvas pack options are set to fill=BOTH  and expand=1.
No, canvases are resizeable.

from Tkinter import *
mFrame = Frame(master=None)
mFrame.pack(fill=BOTH, expand=1)
canvas = Canvas(master=mFrame, bg="red")
canvas.pack(fill=BOTH, expand=1)
mFrame.mainloop()

The tricky part is where and when to use the correct options for the
packer. In your example, the frame was not resizeable, thus the canvas
was not resizeable.

Peter

-- 
---------------------------------------------------------------------------
   Dr. Peter Stoehr --- Teisenbergweg 6 --- 85435 Erding --- 08122/47232
---------------------------------------------------------------------------
		I'm the terror that flaps through the night



From hei at adtranzsig.de  Tue Aug  3 04:43:11 1999
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Tue, 3 Aug 1999 10:43:11 +0200
Subject: Python microthreads
References: <FFropB.7tp@world.std.com>
Message-ID: <933672146.479670@news.adtranzsig.de>

Did you build it on top of Christian Tismers
Stackless Python, or is it a different kind of
modification to the VM?

--
Dirk
Where did you want to go again?

Will Ware schrieb in Nachricht ...
>I have been interested in using Python to program simulated
>swarms of interacting agents. For this purpose, I have implemented
>what I call "microthreads" in Python, modifying Python's virtual
>machine, and maintaining a frame stack for each thread.
[...]





From che at debian.org  Tue Aug 10 15:47:26 1999
From: che at debian.org (Ben Gertzfield)
Date: 10 Aug 1999 12:47:26 -0700
Subject: posix.error is a tuple, not an object or a string as docs say
References: <yttzozzbm11.fsf@gilgamesh.cse.ucsc.edu> <7opulu$15fs$1@nntp6.u.washington.edu>
Message-ID: <yttr9lbbepd.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Donn" == Donn Cave <donn at u.washington.edu> writes:

    Donn> But you're right, it has no "errno" attribute.  I guess this
    Donn> is a version mismatch with your documentation.  If you
    Donn> upgrade to 1.5.2, it will have those attributes - "errno",
    Donn> "strerror" and "filename".  1.5.1's os.error still has only
    Donn> "args".  errno and strerror are args[0] and args[1].

Ah hah. I should have figured it'd be something silly like that.

Thanks.

-- 
Brought to you by the letters K and P and the number 14.
"I'm with insurance."
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From billtut at microsoft.com  Thu Aug 12 15:13:48 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Thu, 12 Aug 1999 19:13:48 GMT
Subject: win2k python install
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB9108F@RED-MSG-50>

Nope. Setup currently leaves you to your own devices with this.

Bill

> -----Original Message-----
> From: Max M. Stalnaker [mailto:stalnaker at acm.org]
> Sent: Thursday, August 12, 1999 10:20 AM
> To: python-list at cwi.nl
> Subject: win2k python install
> 
> 
> Installing python on a win2k RC1 box, I had the expectation that the
> python executable would be on the path.  Thus, from an arbitrary
> directory, with a dos box,  python test.py would find python 
> and execute
> it.  This does not happen and I am wondering if the install is
> *supposed* to put the python executable on the path.
> 
> 



From tchrist at mox.perl.com  Sat Aug 14 12:33:30 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 14 Aug 1999 09:33:30 -0700
Subject: Enough [Was Re: Why use Perl when we've got Python?!]
References: <1277478426-31364719@hypernet.com>
Message-ID: <37b58c4a@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, gmcm at hypernet.com writes:
:John W. Stevens wrote:
:[ crap deleted ]
:
:Whatever your contributions to the python community, you're now in my 
:killfile, along with a handful of pl hotheads who do the same to us.

That man, along with Ian Clarke, have stirred up the biggest flame war
in the Perl newsgroup that we're seen for a long time, inflicting their
standard overrighteous prattle on us.  I realize that they aren't truly
representative of Python people in general: several others have been
entirely reasonable and educative.  I'm not sure hoter users realize
this, however.

On that killfiling issue, I just hope you're still around when I have
Python questions. :-)

--tom
-- 
    : The ksh scripts do not have a problem with it. 
    That's because ksh doesn't much mind opening up security holes.  The
    absence of taint checks is not exactly a feature.
	    Larry Wall in <1994Dec15.013044.16127 at netlabs.com>



From stidolph at origin.ea.com  Thu Aug 19 13:41:07 1999
From: stidolph at origin.ea.com (Stidolph, David)
Date: Thu, 19 Aug 1999 17:41:07 GMT
Subject: How to sell Python to your boss?
Message-ID: <11A17AA2B9EAD111BCEA00A0C9B41793034AAD03@molach.origin.ea.com>

We are building a large client-server system using C++ as the basic
implimentation and Python as a scripting language on top of it.  This gives
us the flexibility of control at any level we want.  Speed issues can be
resolved in C++ and control is kept at the Python level.  When we change
what is going on in the script system we do not have to recompile - just
execute, thus speeding up our development cycle.  In addition, we hope to be
able to make code changes while executing - try that in C or C++!

I started with an extensive background in C/C++ and no knowledge of Python.
I educated myself and after one false start I found SWIG and shadow classes
and just took off.  I'm now a real Python fan in part because of the speed
of development.  I also really appreciate the exception handling system in
Python.  I use it in comparison to NOT using the C++ exception system.

-----Original Message-----
From: Petri Mikael Kuittinen [mailto:eye at niksula.hut.fi]
Sent: Thursday, August 19, 1999 11:54 AM
To: python-list at cwi.nl
Subject: How to sell Python to your boss?


I proposed an interesting idea to my employer for my final thesis.

My proposed final thesis includes:
- a prototype implementation of networked multi-player video gambling
game (which also has a single-player mode)
- a design of a network protocol to allow this and automated testing
of the game (including several automated test player programs which
have different kind of AI). This protocol and test system is designed
so that it can easily be expanded to handle any sort of gambling game

Most of the work is design, documentation and programming a
rudimentary, but easy-to-use, GUI for the multi-player game. My boss
otherwise agreed with my idea ("it would be really useful..."), but he
doesn't want me to do the prototype implementation using Python. He
said I must use a program language that all the other programmers in
our R&D department understand. Well, that leaves only one choice: C

Some of my game programmer colleagues know a little bit of C++, Java
and Perl, but these languages cannot be used either. Operating system
or CPU power is not a problem here (I can use Linux and several
hundred MHz of CPU power).

I have lots of programming experience in C, and much less in
Python. Despite this I am confident that I would be able to implement
the prototype using Python many times faster than using C. Sure I can
do it with C, or even with assembler (I have made a GUI with m68k
asm), but parsing text strings and prototype GUI design is so
laborious in C, that I would only be able to implement a tiny portion
of the whole thing.

I can understand my bosses opinion. He wants to protect the interest
of this firm. If I would leave the firm, then someone else would have
tp spend time learning Python in order to modify my program code.

For several months I have tried to get my colleagues excited to learn
Python, but so far nobody has been interested to learn new languages,
except for some mild interest towards Java (which is IMHO an
overly-hyped language).


Petri

-- 
<(O)> Petri Kuittinen, also known as Eye, Dj Eye or Peku               <(O)>
<(O)> ADDRESS: Timpurinkuja 1 C 39, FIN-02600 Espoo, Finland           <(O)>
<(O)> EMAIL: eye at iki.fi WWW: http://www.iki.fi/~eye/ PHONE: 09-5472380 <(O)>
~The only way to avoid mistakes is to do nothing or try not anything new.



From conv1ct at my-deja.com  Tue Aug 10 08:42:15 1999
From: conv1ct at my-deja.com (conv1ct at my-deja.com)
Date: Tue, 10 Aug 1999 12:42:15 GMT
Subject: any good python source codes i could read?
Message-ID: <7op6n6$7qu$1@nnrp1.deja.com>

hi
i am looking for easy python source codes, or small programs made in
python

convict


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From gmcm at hypernet.com  Mon Aug 30 09:37:46 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 30 Aug 1999 08:37:46 -0500
Subject: Opening files
In-Reply-To: <7qdfld$abj$1@nnrp1.deja.com>
Message-ID: <1276106070-51511659@hypernet.com>

rainer2207 at my-deja.com asks:

> How do I define a function to open a file using
> tkFileDialog.askopenfilename() and what do I have to import?

import tkFileDialog

fnm = tkFileDialog.askopenfilename(
                    filetypes=[("any file", "*")],
                    title="Chose or Lose")

- Gordon



From tim_one at email.msn.com  Fri Aug 20 01:19:05 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 20 Aug 1999 01:19:05 -0400
Subject: sprintf & float (was RE: More random python observations from a perl programmer)
In-Reply-To: <37bc2588@cs.colorado.edu>
Message-ID: <000a01beeacb$85cfc780$712d2399@tim>

[posted & mailed]

[Tom Christiansen]
> ...
>
> 2. Perl code
>
>     main() {
> 	printf("Float %g rounds to %.0f\n", 1.5, 1.5);
> 	printf("Float %g rounds to %.0f\n", 2.5, 2.5);
> 	printf("Float %g rounds to %.0f\n", 3.5, 3.5);
> 	printf("Float %g rounds to %.0f\n", 4.5, 4.5);
>     }
>     main();  # not called by default
>
>     OUTPUT:
>
> 	Float 1.5 rounds to 2
> 	Float 2.5 rounds to 2
> 	Float 3.5 rounds to 4
> 	Float 4.5 rounds to 4


I know Python (and expect Perl) both inherit their sprintf behavior from the
platform libc.  Here's what Perl does under Windows (& Python does too):

C:\temp>perl temp.pl
Float 1.5 rounds to 2
Float 2.5 rounds to 3
Float 3.5 rounds to 4
Float 4.5 rounds to 5

>>>

Unices also vary widely in what they do to 2.5 and 3.5; I don't know of any
platform that doesn't round 1.5 to 2 and 3.5 to 4, though.

add-a-half-&-chop-is-even-older-than-us<wink>-ly y'rs  - tim





From philh at vision25.demon.co.uk  Sat Aug 21 19:56:16 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sat, 21 Aug 99 23:56:16 GMT
Subject: Parrot-0.0.1
Message-ID: <935279776snz@vision25.demon.co.uk>


It looks like version 0.0.1 of Parrot, my text-based GUI builder,
will be ready within a day or so. It doesn't actually do anything
useful yet; v0.0.1 is a technology demonstration version, which
is intended to verify that the basic design is more-or-less
sensible. Would anyone like to have a look at it, and possibly 
suggest some improvements? If so, tell me, and I'll put it on
my website.

Some information about Parrot follows:



An introduction to the Parrot GUI building tool.

Parrot is a text-based GUI builder, written by Philip Hunt. It is 
intended to be used by programmers writing GUI applications. To 
use Parrot, first you create a *.par file describing your 
application's GUI. Your file might look something like this:


-----------Cut here-------------------------------
window @MyWindow "My First Window" {
   menuBar {
      menu "File" { 
         menuItem @New "New" 
         menuItem @Exit "Exit"
      }
      menu "Help" { 
         menuItem @About "About..."
      }
   }
   colLayout {
      rowLayout {
         label "First row!"
         button @Button1 "Press me"
         button @Button2 "And me"
      }
      rowLayout {
         label "Which units:"
         radioButton @inchRB "inches"
         radioButton @feetRB "feet"
         radioButton @yardRB "yards"
         radioButton @mileRB "miles"
      }
   }
}
-----------Cut here-------------------------------


Then you invoke the Parrot executable: this causes Parrot to 
create a file containing source code which, when executed, will 
display your GUI window. 

What language is Parrot's output written in? That depends on 
which _backend_ you use.

Parrot is written in two parts: a frontend, which reads in a
*.par file, like the one shown above, storing it in an 
internal data format; and a backend, which converts the internal
format into code for output.

I intend that multiple backends will be written for Parrot; this
has two advantages:

   1. someone will be able to learn one GUI builder, Parrot, and 
      then use that tool for whatever programming language they 
      are using

   2. a GUI application created in Parrot can be translated into 
      several output formats, e.g. if there is a prototype in 
      one language and the real application in another; or if 
      one is porting to a different OS or GUI toolkit.

Currently, the only backend is a demonstration one that creates
a page of HTML.

Planned backends include:

* Python using the Tkinter GUI library

* the Glade XML format (Glade is a graphical GUI builder which is
  part of the GNOME project; a Glade XML file can be used to build
  GNOME applications).

* C++ using the Qt GUI library

Parrot will also be able to read the Glade XML format and use it
to write *.par files. This means that Parrot should be interoperable 
with Glade -- someone could use the two together to build a GUI.

Anyone wishing to contribute other backends will be welcome to do 
so; hopefully Parrot will eventually encompass all GUI APIs and
prgramming languages in common use. Note that if a GUI toolkit 
doesn't have a visual builder, and someone writes a Parrot backend
for that toolkit, then once the Glade XML->parrot translator has
been written,  


Parrot is written in Python. It is copyrighted by philip Hunt and
licenced under the GNU GPL. Parrot uses version 0.4 of John Aycock's 
parsing framework described in his paper _Compiling Little Languages 
in Python_. 



-- 
Phil Hunt....philh at vision25.demon.co.uk




From guido at cnri.reston.va.us  Thu Aug 19 14:38:40 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 19 Aug 1999 14:38:40 -0400
Subject: CVS classes in Python
References: <37B9969B.BB40C25F@primenet.com> <934914414snz@vision25.demon.co.uk>
Message-ID: <5l7lmrty33.fsf@eric.cnri.reston.va.us>

philh at vision25.demon.co.uk (Phil Hunt) writes:

> In article <37B9969B.BB40C25F at primenet.com>
>            yerny at primenet.com "Bryan Jernigan" writes:
> > I'm a newbie to Python and I would like to create a Python script to
> > maintain a distributed CVS CM system.  Before I re-invent the wheel, I
> > was wondering if anyone has some decent CVS classes and/or functions.
> > If so, please drop me a line.
> 
> Why not have your Python classes acting as a front end to CVS or PVS?

There's also a bunch of old rcs and cvs code in Demo/pdist in the
Python source distribution.

--Guido van Rossum (home page: http://www.python.org/~guido/)



From fredrik at pythonware.com  Tue Aug 31 02:58:42 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 31 Aug 1999 06:58:42 GMT
Subject: Generating an event in Tkinter
References: <7qfc5l$mll$1@nnrp1.deja.com>
Message-ID: <011701bef37e$bcd9ece0$f29b12c2@secret.pythonware.com>

nharlow at my-deja.com wrote:
> I want to generate a dummy event from within a program.  Then bind a
> callback to this event.
> 
> How do I do that?
> 
> In straight tk, it would be something like:
>  event generate parent_window <<foo>>

parent_window.event_generate("<<foo>>")

(additional options can be given as
keyword arguments)

</F>




From john.grayson at gsc.gte.com  Thu Aug 26 06:34:35 1999
From: john.grayson at gsc.gte.com (john.grayson at gsc.gte.com)
Date: Thu, 26 Aug 1999 10:34:35 GMT
Subject: idling with tkinter
In-Reply-To: <7q2u6e$aso$1@nnrp1.deja.com>
References: <7q2u6e$aso$1@nnrp1.deja.com>
Message-ID: <7q357r$c50c@eGroups.com>

john.grayso- at gsc.gte.com wrote: 
original article:http://www.egroups.com/group/python-list/?start=65754
> 
> I found the after_idle() method, but it only calls the callback
> once. Strangely i can't find a method to perminently hook a
> callback into the idle system. Is there one?
> 

You're on the right track. after_idle is call *once* when there are
no pending events in the event queue; you must setup to be called 
*again* before you return from the callback (this is true for after,
too).




From ergowolf at mindspring.com  Sat Aug 21 08:38:46 1999
From: ergowolf at mindspring.com (x)
Date: Sat, 21 Aug 1999 08:38:46 -0400
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu>
Message-ID: <37BE9DD6.CC2B56E3@mindspring.com>

Pythoneers and Tom,
I have been quitely following Tom's posts and it responses over the last several
days and I just wanted to add my two cents.  I have been studying perl for the
past six months or so.  I just bought the MySQL O'Reilley book(finally came
out!) and it said some nice things about Python so I decided to try it out
including listening in on this mailing list and, of Course, the corresponding
O'Reilley, Learning Python.  I must admit the language is interesting and
impressive in the way it works.  I am not an expert in perl or python, but I
like both and they are tools that accomplish very similar tasks.  They are both
interpreted, have links to C, are easy to learn, use Tk,small footprint, etc.
Why learn another language that is so similar?  I think Tom gave some good
reasons.  It seems to me python and perl are competing with each other, because
they are so similar.  I think it would be better to compare the languages to see
how they can be improved rather than taking Tom's posts personally.  I think
taking Tom's posts personally says something about that person, which doesn't do
python justice.  I think if Tom Christiansen takes the time to look at our
language, then that really is a compliment to python, its developers and users.
After all, I have python(I am adding more content as we speak) and perl sections
side by side on my website(link bellow).  Tom I hope you don't get discouraged
and keep posting.  I have learned alot about perl and python from your posts and
the responses.

Ergowolf
www.mindspring.com/~ergowolf
mailto:ergowolf at mindspring.com

PS I would be grateful to anyone who could help me add python content or scripts
to my site.

Tom Christiansen wrote:

> Private mail has indicated a remarkably amount of distrust about my
> learning Python.  Some verges on the unbelievable.
>
> And yes, I react to personal digs unquietly.  That this should
> come as a surprise is itself a surpise.
>
> The first and smaller thing I wish to say is that when I mentioned
> "sifting through the dreck for gems", I meant sorting through the flames
> that have occasionally been levelled at me, to skip their heat and
> continue on in search of useful information on other people's postings.
> I was certainly *not* talking about the language as dreck, as one
> person thought I meant.
>
> Secondly, and much more importantly, the reason I am learning Python
> is *not* because I wish to "score some quick hits and prove to c.l.p.m
> that Perl is best after all" [sic] as one private querent asked, to my
> complete and utter horror.
>
> You may not have known this of me, but I am a collector of languages,
> and to some extent, of language trivia.  Can you honestly believe that
> I studied Spanish, French, German, Italian, Portuguese, or Latin merely
> to prove to myself and the world that English was the best language?
>
> Answer for yourself the question of why I learned those languages,
> and you'll understand my motivations on learning Python.  Think about
> it for a bit, please.
>
> Done thinking already?  Sigh.  Very well, it's quite simple, really:
>
>  *  I want to be able to talk to other people in their own language.
>  *  I want to read their literature.
>  *  I want insights into my own tongue.
>
> Is that really so terribly difficult to identify with?  No conspiracy
> theories of nefarious purposes are required.
>
> Well, that's it for now.  The plane beckons, and I'm off to Monterey.
> I hope to see some of you there.
>
> --tom
> --
> "They always had a rational explanation for the haunting, but they had a dog
>  that could talk."
>                                 - My comment about Scoobie-Doo




From arcege at shore.net  Tue Aug 17 12:00:43 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 17 Aug 1999 16:00:43 GMT
Subject: help printing dictionary, sorted by values.
References: <37B8F03F.654FA9DE@xmission.com>
Message-ID: <LGfu3.394$mV.141087@news.shore.net>

Jerry Williams <jwilliam at xmission.com> wrote:
: I have searched and searched the docs and am really frustrated with
: them.
: All I can seem to find is bits and pieces that hint about dictionaries.
: I finally found a table under mappings that listed the methods.  But
: I couldn't find anything that would list each data type and what methods
: are available.  What I am trying to do is take a file with a bunch of
: entries that appear more than once and list the count of each occurance
: in order of the highest count.  So if the file looked like:
: a
: a
: a
: b
: b
: c
: Then is would print:
: 3 a
: 2 b
: 1 c
: I thought a dictionary would work great, but can't seem to figure out
: how to sort it by values and also get the key.
: list = dict.values()
: list.sort()
: Then I get lost.  And the counts aren't going to be unique.
: Thanks for any help.  Please point me to where the methods are listed
: for
: the dictionary.  And some examples of dictionaries would be helpful.

How about:

  def repr_sorted_dict(d):
    """Return a string representation of a dictionary,
  sorted by value first."""
    def cmpitems_by_value(a, b):
      """sort first by second element in sequence."""
      v = cmp(a[1], b[1])
      # if equal, then sort by first element in sequence
      if v == 0:
        v = cmp(a[0], b[0])
      return v
    import string
    # retrieve the (key, value) pairs
    items = d.items()
    items.sort(cmpitems_by_value)
    # display the pairs
    l = []
    for k, v in items:
      l.append("%s: %s" % (`k`, `v`))
    # now combine them as a dictionary display
    return '{' + string.joinfields(l, ', ') + '}'

Then you can:
  print repr_sorted_dict(my_values)

And if you really wanted to get fancy, you could make this function the
__repr__ method of a subclass of UserDict.UserDict.

  from UserDict import UserDict
  class ReverseSortedDict(UserDict):
    __repr__ = lambda self: repr_sorted_dict(self.data)

(Although I'd rewrite it to use the UserDict interface rather than
self.data itself, but that's just me. :)

Even if this is not exactly what you want, the "pairs = dict.items();
pairs.sort(...)" should get you where you are going. :)

This is documented at:
  http://www.python.org/doc/current/lib/typesmapping.html
  http://www.python.org/doc/current/lib/typesseq-mutable.html

  -Arcege




From tchrist at jhereg.perl.com  Thu Aug 19 14:19:00 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Thu, 19 Aug 1999 12:19:00 -0600
Subject: More random python observations from a perl programmer 
In-Reply-To: Jeremy Hylton's message of Thu, 19 Aug 1999 14:20:03 -0400 (EDT) .
             <14268.19155.505284.821500@bitdiddle.cnri.reston.va.us> 
Message-ID: <199908191819.MAA04074@jhereg.perl.com>

>Yup.  It says that Windows help is useful for more people than icky
>manpages.  

No, it means that it's considered ok to alienate people who expect manpages.

--tom



From singhbhattig at my-deja.com  Tue Aug 24 21:13:18 1999
From: singhbhattig at my-deja.com (singhbhattig at my-deja.com)
Date: Wed, 25 Aug 1999 01:13:18 GMT
Subject: equivalent of "getch" of C
Message-ID: <7pvfv6$s01$1@nnrp1.deja.com>

Hello,

What is equivalent of getch in jpython, if any.

thanks

GSB


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From chad at vision.arc.nasa.gov  Mon Aug 23 16:49:01 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Mon, 23 Aug 1999 20:49:01 +0000
Subject: ANN: PmwContribD 1.11 - Ahhhh nuts!
References: <7prk5t$2ed$1@nnrp1.deja.com>
Message-ID: <37C1B3BD.9FB65A10@vision.arc.nasa.gov>

Doug Hellmann wrote:

> This library consists of several Pmw-based Tkinter widgets including a
> ProgressMeter, basic TreeNavigator and a GUI application framework for
> creating simple GUI apps quickly.

   Man, I just finished writing a progress meter widget for an application
that I'm writing...  Oh well. :-)  Thanks for the pointer, I like yours and
may use it instead.

Cheers,
Chad
chad at vision.arc.nasa.gov




From def-p at usa.net  Mon Aug 30 17:45:14 1999
From: def-p at usa.net (Def P)
Date: Mon, 30 Aug 1999 21:45:14 GMT
Subject: Grid (possible scrollable) in Tkinter?
Message-ID: <19990830214514.12034.qmail@www0j.netaddress.usa.net>

Hello,

I'm looking for something like a grid in Tkinter... actually I do not want to
borrow one from existing libraries, but build one myself from Tkinter pieces,
if possible. Is this a reasonable thought? :) Would I be able to make the grid
scrollable (easily)? 

Another pro^H^H^Hchallenge: I noticed/think that Tkinter does not have a
scrollable listbox. To make one isn't that hard, just throw a Listbox and a
Scrollbar together in a Frame and link them. I would like to add some more
features, like typing a letter to jump to the first item in the list starting
with that letter. ...Is building these things worthwile or shouldn't I bother
and look into libraries like Pmw? For me, the ideal situation would be if I
could distribute my programs without requiring external libraries.

TIA,
Def P


____________________________________________________________________
Get free email and a permanent address at http://www.netaddress.com/?N=1



From bwarsaw at cnri.reston.va.us  Mon Aug 16 18:12:16 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Mon, 16 Aug 1999 18:12:16 -0400 (EDT)
Subject: Python CGI Scripts
References: <Pine.LNX.4.04.9908162019270.3026-100000@natal.stamp-coin.com>
	<slrn7rgucs.udq.wtanksle@dolphin.openprojects.net>
Message-ID: <14264.36032.260071.711968@anthem.cnri.reston.va.us>

>>>>> "WT" == William Tanksley <wtanksle at dolphin.openprojects.net> writes:

    WT> I'm pretty sure it's already there, even for people who don't
    WT> use Bobo.  People have been using Python for CGI scripting for
    WT> a while.

Take a look at Mailman <www.list.org>.  It is in large part a huge
Python CGI `script'!

-Barry



From skip at mojam.com  Fri Aug 20 01:10:07 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 20 Aug 1999 00:10:07 -0500 (CDT)
Subject: No man pages? (was Re: More random python observations from a perl programmer)
In-Reply-To: <37BCD4A5.482CC246@montana.com>
References: <199908192043.OAA07181@jhereg.perl.com>
	<37BCD4A5.482CC246@montana.com>
Message-ID: <14268.58085.361375.781012@dolphin.mojam.com>

    bowman> You want to be careful building Python 1.5.2, also, as a little
    bowman> misstep will break RH's 'glint', 'linuxconfig' and other Python
    bowman> based oddities.

There are two or three scripts that need to have their function def's fixed
to not have default args after non-default args.  If anyone needs them I can
build a context diff. Just let me know.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From crdempsey at uswest.net  Wed Aug 25 08:00:57 1999
From: crdempsey at uswest.net (Syniq)
Date: Wed, 25 Aug 1999 06:00:57 -0600
Subject: Small question about Python executeable(sp?) files
Message-ID: <37C3DAF9.D51DAAF5@uswest.net>

I know that

#! /something/blah/whatever/python

as the first line in your code will make your "program" an
executeable(sp?), but this assumes that your user has python installed
in /something/blah/whatever/python.

What I'm wondering: Is there a way to make the script execute regardless
of the location (or even existence) of python on the user's machine
(basically, compile it into something sort of resembling ye old .EXE
file)?


Syniq
crdempsey at uswest.net




From tim_one at email.msn.com  Mon Aug 23 02:06:59 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 23 Aug 1999 02:06:59 -0400
Subject: matching any number of any character
In-Reply-To: <wr34shrtkk1.fsf@gargoyle.cs.uchicago.edu>
Message-ID: <000501beed2d$b5c74800$352d2399@tim>

[Lyn A Headley]
> To do it, I want to use the regex module, heretical though that may
> be.

Not heretical so much as ornery and self-destructive <wink>.

> I think my brain is missing a piece, because I can't seem to
> accomplish such a task.  I thought this would work:
>
> >>> import regex
> >>> rx = regex.compile('\(.\|\n\)+')
> >>> rx.match('abc')
> 3
> >>  rx.group(1)
> 'c'
>
> but I wanted 'abc'!

Persist:

>>> rx.group(0)
'abc'
>>>

The thing inside your parens only matches one character; repeating it doesn't
change that; you get back the character (c) it matched last.  Life is easier
with re:

>>> import re
>>> rx = re.compile("(.+)", re.DOTALL)
>>> m = rx.match("abc")
>>> m.group(1)
'abc'
>>>

this-is-your-brain-that-was-your-brain-on-regex-ly y'rs  - tim





From proenca at hks.com  Tue Aug 17 11:46:14 1999
From: proenca at hks.com (Pedro Proenca)
Date: Tue, 17 Aug 1999 11:46:14 -0400
Subject: ANN: PythonWare Sound Toolkit (PST) release 0.1
References: <007d01bee8b7$6ecbc5e0$f29b12c2@secret.pythonware.com>
Message-ID: <37B983C6.48E6BEAD@hks.com>

PythonWare wrote:
> 
> The folks who gave you the Python Imaging Library (PIL) now
> introduces yet another useful library:
> 
> The PythonWare Sound Toolkit (PST) reads sound files in a
> variety of formats, and plays them on different hardware plat-
> forms (currently Windows and Sun only).
> 
> The first, very rough, release is available from:
> 
>     http://www.pythonware.com/
> 
> PST 0.1 is shipped under a Python-style license, and can be used
> without fee also in commercial projects.  For more info, see:
> 
>    http://www.pythonware.com/products/pst/
> 
> enjoy,
> the PST development team


How come it doesn't support MP3?

-- 
Pedro Proenca
---------------
proenca AT hks DOT com

"I love deadlines. I especially like the whooshing sound they make as they go
flying by." -- Douglas Adams



From gvwilson at nevex.com  Sat Aug 28 12:24:32 1999
From: gvwilson at nevex.com (Greg Wilson)
Date: Sat, 28 Aug 1999 12:24:32 -0400
Subject: echoing input
Message-ID: <7q92j8$naa$1@demon.uunet.ca>

Hi.  Is there a way to get Python to echo the contents of a script while it
is
being loaded?  I have a script called x.py that contains:

# x.py
a = (1, 2, 3)
print a

and I would like to see

a = (1, 2, 3)
print a
(1, 2, 3)

as the output from "python x.py".

Thanks,
Greg





From vitus at wagner.rinet.ru  Mon Aug 23 14:42:18 1999
From: vitus at wagner.rinet.ru (Victor Wagner)
Date: 23 Aug 1999 22:42:18 +0400
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk>
Message-ID: <7ps4ma$n06$1@wagner.wagner.home>

Phil Hunt <philh at vision25.demon.co.uk> wrote:


: Parrot is a text-based GUI builder, written by Philip Hunt. It is 
: intended to be used by programmers writing GUI applications. To 
: use Parrot, first you create a *.par file describing your 
: application's GUI. Your file might look something like this:

Compare this with code below:
: -----------Cut here-------------------------------
: window @MyWindow "My First Window" {
:    menuBar {
:       menu "File" { 
:          menuItem @New "New" 
:          menuItem @Exit "Exit"
:       }
:       menu "Help" { 
:          menuItem @About "About..."
:       }
:    }
:    colLayout {
:       rowLayout {
:          label "First row!"
:          button @Button1 "Press me"
:          button @Button2 "And me"
:       }
:       rowLayout {
:          label "Which units:"
:          radioButton @inchRB "inches"
:          radioButton @feetRB "feet"
:          radioButton @yardRB "yards"
:          radioButton @mileRB "miles"
:       }
:    }
: }
: -----------Cut here-------------------------------


toplevel .mywindow -menu .mywindow.m; wm title .mywindow "My First Window"
menu .mywindow.m
  .mywindow.m add cascade -label "File" -menu .mywindow.m.file 
     menu .mywindow.m.file
        .mywindow.m.file  add command -label New -command new_file
        .mywindow.m.file add command -label Exit -command exit
  .mywindow.m add cascade -label "Help" -menu .mywindow.m.help
     menu .mywindow.m.help
        .mywindow.m.help add command -label About... -command about
frame .mywindow.f1
  label .mywindow.f1.l1 -text "First row"
  button .mywindow.f1.b1 -text "Press me" -command do_something
  button .mywindow.f1.b2 -text "And me" -command do_something_else
  pack .mywindow.f1.l1 .mywindow.f1.b1 .mywindow.f1.b2 -side left -padx 10
  
frame .mywindow.f2
 label .mywindow.f2.l1 -text "Which units:"
 # Here I got tired of typing and let the program produce rest
 pack .mywindow.f2.l1 -side left -padx 10
 foreach unit {inches feet yards miles} {
    radiobutton .mywindow.f2.$unit -text $unit -value $unit -variable unit
    pack .mywindow.f2.$unit -side left -padx 10
 }
pack .mywindow.f1 .mywindow.f2 -side top -pady 5 -anchor n

This is tcl/tk. Almost no difference. A bit more verbose, but there are
numerous ways to shorten code, which I haven't used intentionally until
I got tired and used a loop to produce four radiobuttons.

And this is not something which should be run through preprocessor.
This is already runnable code. Just add #!/usr/bin/wish at the beginning
and chmod +x it. And if you define procedures do_something ,
do_something_else and new_file, it may even do something useful.
Oh yes, there is empty default window floating around. I could create
all the widgets in it directly, but in original code it was obvoisly not
main window of program, so I choose to create separate toplevel.

Note also that all this things are done runtime, not when preprocessing,
so, for instance I could get list of units via database query or
substitute different lists according to current locale.

Once again: who need so-called GUI builders while we have scripting
languages, which provide same level of abstraction, but much more
flexibilty, becouse interface is created run-time? 

Really it is very simple to define parrot interpreter on tcl - just
define commands such as window, columnLayout, rowLayout as tcl procedures 
which substitute
something by default. And parrot files would become runnable. But pretty
useless, becouse I've not seen any way to attach functionality to these
windows.


-- 
--------------------------------------------------------
Victor Wagner @ home       =         vitus at wagner.rinet.ru 
I don't answer questions by private E-Mail from this address.



From skaller at maxtal.com.au  Sun Aug 29 23:43:26 1999
From: skaller at maxtal.com.au (John Skaller)
Date: Mon, 30 Aug 1999 13:43:26 +1000
Subject: [Types-sig] Re: Python Type System: An idea for unification
In-Reply-To: <1276189173-46513020@hypernet.com>
References: <3.0.6.32.19990829154836.009664b0@mail.triode.net.au>
 <1276273608-41434292@hypernet.com>
Message-ID: <3.0.6.32.19990830134326.00967540@mail.triode.net.au>

At 09:33 29/08/99 -0500, Gordon McMillan wrote:
>John (Max) Skaller wrote:
>> At 10:05 28/08/99 -0500, Gordon McMillan wrote:
>> >John (Max) Skaller wrote:
>
>> > [Briefly, and subject to my misinterpretation, it appears that
>> >  Viper objects always have a type obj; said type obj may be
>> >  a class instance; said type obj is appended to the normal
>> >  attribute search path]
>> 
>>  "Usually" the type object will be a class, rather than
>> an instance.
>
>*That's* what I was missing.

	Yeah: it's more general than just 'a type is a class'.
A type can be any object.

> The changes from stock Python are:
>   1) Tweaking attribute lookup (appears to be a simplification!).

	If it were implemented in CPython, it might require a bit more
code. I suspect this can be implemented in CPython, quite easily,
but I'm not sure. 

>   2) How objects are associated to types.
>      - builtins have a key that associates them to their type
>        objects thru a global dict

	I consider this a hack: the indirection is forced
by the way the interpreter initialises. There is probably
a better technique, such as making the object --> type
link a "pointer to pointer" intead of a pointer,
(a tiny overhead), and then setting the type to 'not available yet'
at startup. Then, after the py_types.py module is loaded,
the types can be set to what was defined in that module:
that is, the __typename__ lookup is done, but exactly once
per type, not once per lookup per object.

	This would slow down lookup by exactly one dereference,
which is trivial. It also allows 'global' instrumentation of
a type by dynamically changing the type of all objects
of a given type to some other type.

>      - users can create objects of arbitrary type thru a new
>        builtin "new_object(type_object)"

	That is the idea: but note it is not implemented yet.

>      - extension writers have a bit more work to do, in that 
>        something more sophisticated than a static C struct is
>        required.

	This is only one third true <g>. First, I am currently 
an extension writer, and it is easy to build extensions,
because the native language is Ocaml, not C. Ocaml is MUCH
easier to write than C! And many of the extensions required
have already be implemented in Ocaml: for example unix
stuff like sockets, files, processes; and just recently
a PCRE (Perl style regular expressions).

	Second, many extensions being written today in C
would not be required if the compiler turns out to work
well enough: Python will do instead of C.

	Finally, Ocaml supports C extensions. These would be
a bit harder to write, since you have to interface
Viper to ocaml wrappers of the C functions, which would
have to support the ocaml interface. 

	HOWEVER, the job is made easier because you don't
need to worry about methods, lookup, or other stuff.
Instead, you just implement functions, and put them
into a dictionary, and then you write PYTHON script
to use them to model an object.

	This is what I am doing for files.
The native file object is built into the interpreter at
the moment and there are a few functions like
'file_open', 'file_read', 'file_close'. These functions
are then called by class methods of the PyFileType
object, and the builtin 'open' function is a Python
function which creates an instance of it.

	So, for example, 'readline' is actually
implemented in Python: it reads some bytes, usin 'file_read'
uses 'string.find' to look for a newline,
and returns the string up to the newline.

	[Actually, I do something more complex: there are
TWO file types: PyNativeFileType and PyFileType. The first
is a class which is a lightweight wrapper around the
builtin file access functions, it is the type object
for the native file type (which is not a class instance).

	The second object is a class, and the usual
'pyhthon file object' is actually an instance of it:
one of the attributes is a PyNativeFileType object.
In other words, 

	print type(open("file"))
	assert type(open("file")) is type(SomeClass())

will print

	<Instance of PyFileType>

and you can derive a new class from this class,
and set attributes:

	f = open("filename")
	f.my_attr = "An opened file"

because it is a normal instance object.

>Is that a fair summary?

	Yes.

>Questions:
>
> Do the type objects of modules, functions, tracebacks... (where the 
> new flexibility would not appear useful) follow these rules?

	The type objects of modules, functions, etc, are all
normal python class objects at the moment. However,
non-method lookup is _not_ done generically. For example:

	x + y

Here, the interpreter does a typecase on x and y, and calls the
appropriate function. So you can't 'hook' the + operator
with a __add__ method in the type objects for x and y
if they're integers, for example. 

	The only lookup that is 'hooked' at the moment is the notation:

	object.attribute # used as an rvalue
	getattr(object,"attribute")

> Have you thought at all about the current metaclass hook? 

	Not especially. The metaclass hack <g> is an implementation
artefact discovered after the fact. I'm interested in 
meta-programming, but I think it is better to design
the architecture to support it from the ground up,
rather than relying on a design accident. :-)

	If I understand correctly, the metaclass hook comes
about when the base of a class is not a class. At present,
Viper throws an exception when this is the case. 

	I will investigate allowing bases to be 'any object': 
I can't see why it shouldn't work, however there is ticklish
lookup problem here: the 'type object' lookup on
an instance of a class only happens if the 'usual'
lookup fails. The usual lookup searches the class
of an instance and its bases, after searching the
attribute dictionary.

	The question is: what happens while searching the class and bases?
Normally, this is just a lookup in the class dictionary:  I can't
remember if the lookup in the type object PyClassType happens,
because that class is defined by:

	class PyClassType: pass

i.e. the lookup wouldn't find anything anyhow. I'm not sure that
a 'full' lookup in the bases would make sense, because it would
happen _before_ the lookup in the type object of the _instance_.
The result would be a type method of the type of a base,
rather than the type of the instance: surely the type of the
instance should be searched first?

	I.e. the algorithm should be:

	1. look in the instance dictionary
	2. look in the class dictionary, and its bases,
		recursively
	3. look in the type of the instance.

Note that step 2 does not look in the class, it looks
in the class dictionary. This is a different thing,
since looking in the class implies looking in the
class type object immediately the looking in the
dictionary fails.

	The distinction is subtle and currently
irrelevant. However, step 2 does NOT make sense
if bases are not classes. So step 2 would have to
become a generalised lookup.

	Just to blow your brain: remember that
if the TYPE object is an instance, the whole
lookup recurses: we not only look in the type
object instance dictionary, but then also
in the class of the instance, and then
in the type of the type!!!!

>That is,
> a way of creating class instances where the type is something other
> than the "stock" InstanceType? 

>or perhaps, the type of the 
> instance's class is something other than the "stock" ClassType?

	If I provide a 'settype(object, typeobject)'
method, you could change the type object of any object
to be any other object. 

	Another possibility is that, like the
__getattr__ method of a class, you could hook type fetching with
a __gettype__ method, and another possibility is to provide
explicit control over the lookup algorithm(s) (globally).

	The thing I need to keep in mind, though, is that
I want to _compile_ python code. So support for
'extreme dynamism' is likely to be restricted to 
client constructed objects: I can't have clients
fiddling things like integer addition, because the
compiler needs to 'know' how to do that, to generate
optimal code.

	Similarly, it would be nice to make _some_
Python classes onto the moral equivalent of the good
old C struct, for high speed lookup; i.e. eliminate
the overhead of a dictionary lookup.

>listening-for-the-sound-of-a-down-under-brain-exploding-ly y'rs

	:-)

-------------------------------------------------------
John Skaller    email: skaller at maxtal.com.au
		http://www.maxtal.com.au/~skaller
		phone: 61-2-96600850
		snail: 10/1 Toxteth Rd, Glebe NSW 2037, Australia





From skaller at maxtal.com.au  Fri Aug 27 20:34:22 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sat, 28 Aug 1999 00:34:22 GMT
Subject: RFI: What changes in Python 1.6
Message-ID: <37c72dee.1411808533@news.triode.net.au>

Would someone with access to Python 1.6 working sources
like to summarise changes from Python 1.5.2 for me
and others that might be interested?

I'm particularly interested in any new methods objects
support, new builtin functions, and of course architectural
changes.


John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From bowman at montana.com  Thu Aug 19 08:49:04 1999
From: bowman at montana.com (bowman)
Date: Thu, 19 Aug 1999 06:49:04 -0600
Subject: PIGgies?
References: <19990818171523.A22207@digex.net> <37BB6216.55DC0E79@rubic.com>
Message-ID: <37BBFD40.449C8C1E@montana.com>

Jeff Bauer wrote:
> 
> PIGgies take their sense of humor seriously.  Don't mess
> with them unless you want a pie in your face.  ;-)

Do little PIGgies grow up to become cans of Spam??

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From ze5yr at my-deja.com  Fri Aug 13 19:07:04 1999
From: ze5yr at my-deja.com (ze5yr)
Date: Fri, 13 Aug 1999 23:07:04 GMT
Subject: Interesting thread on comp.lang.perl.misc
References: <37B42034.33F77AD5@strs.co.uk> <v9Ws3.1713$EG4.332188@news.shore.net> <Pine.GSO.4.10.9908131047510.3761-100000@crusoe.crusoe.net>
Message-ID: <7p28ei$l7i$1@nnrp1.deja.com>

In article <Pine.GSO.4.10.9908131047510.3761-100000 at crusoe.crusoe.net>,
  japhy at pobox.com wrote:
> I wish I could change a string in-place with a function.*
> [bobbit]
> * meaning:
>
> def func(str):
>   # something modifying str
>
> str = "foo"
> func(str)
> print str # prints... "bar", for example

Enclose it in a list, which is mutable:

def func(l):
  l[0]="bar"

>>> l=["foo"]
>>> func(l)
>>> print l
['bar']

A little clumsy, but it works!


--
Cliff Crawford                           -><-
http://www.people.cornell.edu/pages/cjc26
                            empty-handed I go,
)O(       and behold the spade is in my hands


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kaz at ashi.FootPrints.net  Tue Aug 24 22:16:56 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Wed, 25 Aug 1999 02:16:56 GMT
Subject: equivalent of "getch" of C
References: <7pvfv6$s01$1@nnrp1.deja.com>
Message-ID: <slrn7s6khl.53t.kaz@ashi.FootPrints.net>

On Wed, 25 Aug 1999 01:13:18 GMT, singhbhattig at my-deja.com
<singhbhattig at my-deja.com> wrote:
>Hello,
>
>What is equivalent of getch in jpython, if any.

Doh, C has no getch function. This is a platform-specific extension
supported by some C compilers.

Specifically, the libraries that come with some MS-DOS compilers have a getch()
function that is declared in the <conio.h> header which is specific to that
environment.

There is also a getch() function in the curses library.



From larsga at ifi.uio.no  Wed Aug  4 13:50:42 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 04 Aug 1999 19:50:42 +0200
Subject: "hello, world" of embedding/extending?
References: <7o9sum$2ak$1@news.panix.com>
Message-ID: <wkk8rb4eq5.fsf@ifi.uio.no>

* kj0 at mailcity.com
|
| What are the "hello, world" equivalents of embedding and extending
| Python (in/with C or C++)?

Look here:
<URL: http://www.python.org/doc/current/ext/ext.html >

--Lars M.



From JamesL at Lugoj.Com  Fri Aug 27 01:23:31 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Thu, 26 Aug 1999 22:23:31 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net> <37C552DF.9F90BA36@megsinet.net> <37c563bc.26497369@news.hydro.co.uk> <935699430snz@vision25.demon.co.uk>
Message-ID: <37C620D3.A47C5BA@Lugoj.Com>

Phil Hunt wrote:
> 
> In article <37c563bc.26497369 at news.hydro.co.uk>
>            gary at onegoodidea.com "Gary O'Keefe" writes:
> > Perl on the other hand is the first language anyone even vaguely
> > interested in putting together an interactive web-server hears about.
> > That means EVERY DUMBASS IN THE UNIVERSE who wants to put their cock
> > on the internet and have you virtually stroke it but, importantly,
> > can't be bothered to actually learn all this computer bullshit posts
> > to comp.lang.perl.misc.
> >
> > If your newsgroup (that you'd just paid to download) was filled with
> > posts like this:
> >
> > > Subject: [newbee] I am too damn lazy - can't you do this for me?
> > >
> > > hay gurus
> > >
> > > Im starting up an intractive cock-stroking website
> > > (http://www.cockstroke.com/). Its going 2 B KEWL!!!
> > >
> > > But I NEED A CGI TO MAKE IT WORK!!!
> > >
> > > Can you supr-clever gurus write me something??!! It woudnt take
> > > you any time and ITD B GRATE!!!
> > >
> > > THX
> > >
> > > Newbee
> >
> > then your patience would vapourise, just like ours.
> I can see it now: Python, the language for a firmer, harder, hotter,
> cock-stroking experience!!!
> 

Please Gentlemen. These comments should be directed to the comp.lang.viagra
newsgroup. Only the Viagra language is capable of providing macho
programmers with the stamina to satisfy their computer's lust for more
things to compute. Or perhaps it is best that we not pursue this line of
inquiry....



From aahz at netcom.com  Sun Aug 22 10:46:12 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Aug 1999 14:46:12 GMT
Subject: [Off-topic] Sysadmins (was Re: can some one help a newbie?)
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <935227166snz@vision25.demon.co.uk> <7pmv4j$gtt@dfw-ixnews14.ix.netcom.com> <935273107snz@vision25.demon.co.uk>
Message-ID: <7pp2fk$7ul@dfw-ixnews9.ix.netcom.com>

In article <935273107snz at vision25.demon.co.uk>,
Phil Hunt <philh at vision25.demon.co.uk> wrote:
>In article <7pmv4j$gtt at dfw-ixnews14.ix.netcom.com>
>           aahz at netcom.com "Aahz Maruch" writes:
>> In article <935227166snz at vision25.demon.co.uk>,
>> Phil Hunt <philh at vision25.demon.co.uk> wrote:
>>>In article <7pkke6$lut at dfw-ixnews6.ix.netcom.com>
>>>           aahz at netcom.com "Aahz Maruch" writes:
>>>>
>>>> Overall, if your background is as a sysadmin using shell scripts, grep,
>>>> awk, sed, and so on, you'll find Perl a bit easier.
>>>
>>>I've used all these, and find Python easier than Perl. Perl would
>>>probably be easier than Python for 1-liners, if I could remember
>>>the syntax.
>> 
>> Note that I'm specifically assuming a sysadmin with little actual
>> programming experience.
>
>The idea that a company would employ someone, who can't really code,
>as a sysadmin is IMO frightening. Unless they don't care whether
>their computer system works, of course.

I guess I'm missing something, because the prospect doesn't "frighten"
me.  There are a lot of activities for which a little bit of programming
goes a long way, but I've seen a lot of people capable of doing that who
really are not programmers in any sense of the word that you and I would
use.  That's precisely a large part of why Perl has been so successful;
it lends itself readily to the kind of "cookbook" programming that
sysadmins do.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From philh at vision25.demon.co.uk  Sun Aug  8 08:31:22 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 08 Aug 99 12:31:22 GMT
Subject: An idea - request for comments
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com> <37AA804D.9A1317C0@houbits.com>
Message-ID: <934115482snz@vision25.demon.co.uk>

In article <37AA804D.9A1317C0 at houbits.com> nhead at houbits.com "Nigel Head" writes:
> There's no licence statement in the distribution at all for now --
> my intention is that anyone can use the contents anyway they want,
> for non-commercial purposes, providing they keep a reference to the
> original source. Oh yes, and I can't take any responsibility for errors.
> 
> Perhaps one of you experienced people can offer some quick advice as to what I
> need to do -- do I need to just add a licence.txt file or must I add a header
> to every file? What standard text can I use? Is my intent compatible with the
> GPL?

No. The GPL allows anyone to use the software, for commercial or
non-commercial purposes. However, if they distribute the software, 
they must include all source code, including all source code they
write which is linked to the software.

The license that you want is something like the BSD license except that
it forbids commercial use.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From aahz at netcom.com  Mon Aug 30 14:05:06 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 30 Aug 1999 18:05:06 GMT
Subject: Library philosophy
References: <37C57A80.D5C62704@prescod.net> <935699892snz@vision25.demon.co.uk>
Message-ID: <7qeh4i$bg5@dfw-ixnews4.ix.netcom.com>

In article <935699892snz at vision25.demon.co.uk>,
Phil Hunt <philh at vision25.demon.co.uk> wrote:
>In article <37C57A80.D5C62704 at prescod.net>
>           paul at prescod.net "Paul Prescod" writes:
>>
>> Would it it make sense to start wrapping the low-level modules with
>> high-level, easier to use modules that would be "the default" in Python
>> 2?
>
>This could be a useful idea, as long as the present libraries are
>kept for compatibility.
>
>I've written a class representing a Usenet article, which could form
>the basis of a high-level nntp module.

Interesting.  Your second paragraph proves your first, because netnews
articles per se comprise only part of the NNTP spec.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From cjw at connection.com  Tue Aug 31 09:03:16 1999
From: cjw at connection.com (Colin J. Williams)
Date: Tue, 31 Aug 1999 09:03:16 -0400
Subject: Function calling problem
References: <7qb06p$mj5$1@nnrp1.deja.com>
Message-ID: <37CBD294.D79F6EA1@connection.com>

I've made a couple of changes.  I hope this helps.

Colin W.

'''
From:
          rainer2207 at my-deja.com

Sun 5:57 AM

  Subject:
          Function calling problem




Can someone please explain what's wrong with this code? I've tried to
call the function quit(self) but I get an error. What's wrong here?
'''

from Tkinter import *
import sys, Dialog

class Interface:


        def __init__(self, master):

                self.menu = Menu(master, tearoff=0, name="menu")
                master.configure(menu=self.menu)

                self.frame = Frame(master, width="15c", height="10c",
bd=2, relief="sunken")
                self.frame.pack(fill="both", expand=1)

                m = Menu(self.menu, tearoff=0)
                self.menu.add_cascade(label="File", menu=m, underline=0)

                n = Menu(m, tearoff=0)
                m.add_cascade(label="New", menu=n, underline=0)
                n.add_command(label="RDF Schema", underline=4)
                n.add_command(label="RDF Metadata", underline=4)

                o = Menu(m, tearoff=0)
                m.add_cascade(label="Open", menu=n, underline=0)
                o.add_command(label="RDF Schema", underline=5)
                o.add_command(label="RDF Metadata", underline=5,
                   command=self.open)

                m.add_command(label="Save", underline=0)
                m.add_command(label="Save As...", underline=5)
                m.add_command(label="Close", underline=0)
                m.add_separator()
                m.add_command(label="Exit", underline=0,
                  command=self.quit)

                m = Menu(m, tearoff=0)
                self.menu.add_cascade(label="Edit", menu=m, underline=0)
                m.add_command(label="Cut", underline=2)
                m.add_command(label="Copy", underline=0)
                m.add_command(label="Paste", underline=0)

                m = Menu(m, tearoff=0)
                self.menu.add_cascade(label="Help", menu=m, underline=0)
                a = Menu(m, tearoff=0)
                m.add_command(label="About", underline=0)

        def quit(self):
                print 'Exit'
                root.destroy()                                     #
*************
                sys.exit(0)

        def open(self):                                            #
************
          print 'What is this supposed to do?'

root = Tk()
inter = Interface(root)
root.title("Title")
root.mainloop()









From tjreedy at udel.edu  Mon Aug 23 14:39:37 1999
From: tjreedy at udel.edu (Terry Reedy)
Date: 23 Aug 1999 18:39:37 GMT
Subject: filtering a list
References: <yILNBDAxDVw3EwSE@jessikat.demon.co.uk>
Message-ID: <7ps4h9$c14$1@news.udel.edu>

In article <yILNBDAxDVw3EwSE at jessikat.demon.co.uk>, 
robin at jessikat.demon.co.uk says...
[posted and emailed]

>I'm uncertain as to whether this kind of construct is well formed
>R is a list of instances
>
>for i in R:
>  if i.id%2: del R[R.index(i)]

I believe that modifying the for-list is unwise since i will never be 
pointed at item after that deleted when its reference is shifted down 
into now empty slot.  IE, 'for' creates internal index but not internal 
copy.

>or is it better (assuming R holds the only references to the instances)
>to do
>
>R=filter((lambda x: x.id%2),R)
>-- 
>Robin Becker

In this case yes.
Terry J. Reedy




From kernr at mail.ncifcrf.gov  Fri Aug 20 01:10:57 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Fri, 20 Aug 1999 05:10:57 GMT
Subject: Python to Perl Conversions
References: <1276999045-22443458@hypernet.com>
Message-ID: <37bce250.48509376@news.erols.com>

On Fri, 20 Aug 1999 00:27:48 -0500, "Gordon McMillan"
<gmcm at hypernet.com> wrote:

>Couple comments -
>
>Tom Christiansen wrote:
>>                                 Yes, slices in python don't include
>>                                 the termination point.
>
>This might be a surprise if you're not used to it, but it turns out 
>to be very nice. Basically any way of expressing a range will result 
>in something that includes the left hand side, but excludes the right 
>hand. So when you're slicing, chopping up, inserting... you (almost) 
>never have to use "i-1".

It also means that len(seq[start:end]) == end - start.  It is also
convenient when using negative indices.

>> delattr(object,name)            n/a?  "delete named attribute from
>>                                 object".
>>                                 I suppose this might be
>>                                         delete $obj->{name}
>>                                 but that's pretty uncool to do it
>>                                 without asking the object.
>
>I'm not sure I've *ever* seen this used in Python. But it hints at a 
>very nice piece of syntactic sugaring you can do.

I've seen it used in wrapper classes where the classes' __delattr__
method needs to delete the data object's attribute.

[snip rest]

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From jhspies at futurenet.co.za  Thu Aug  5 08:41:35 1999
From: jhspies at futurenet.co.za (Johann Spies)
Date: Thu, 5 Aug 1999 14:41:35 +0200 (SAST)
Subject: Printing from python on Windows platform
In-Reply-To: <C12567C3.003009B3.00@derumg01.cyberlink.eds.com>
Message-ID: <Pine.LNX.3.96.990805143604.565B-100000@Johann>

On Wed, 4 Aug 1999 david.stegbauer at cz.opel.com wrote:


> Printing under windows is not such easy. I personally write to a file,
> then I pass it to MS Word to print. I format the output using RTF, but I
> think there is no problem to use HTML and Netscape - then it is
> sufficient to spawn it like
> 
> from os import spawnv, P_NOWAIT spawnv(P_NOWAIT,
> r'D:\APP\NETSCAPE\PROGRAM\NETSCAPE.EXE', ('NETSCAPE.EXE',
> '/print("filename.html")')) 

That is quite complicated.  I do not have MS Word and do not know the file
structure of RTF or HTML at the moment.  I suppose to go this route it
will be easier for me to use LateX for formatted output.  After some
suggestions sent to me I am using l = open('LPT1',w) and l.write and it
works for just straight forward printing.

Thanks for your input.  As somebody who do not know windows, this is quite
new to me and I will experiment with your code to try and understand what
is happening.

Johann.
 --------------------------------------------------------------------------
| Johann Spies                                 Windsorlaan 19              |
| jhspies at futurenet.co.za                3201 Pietermaritzburg             |
| Tel/Faks Nr. +27 331-46-1310		       Suid-Afrika (South Africa)  |
 --------------------------------------------------------------------------

     "I am crucified with Christ, nevertheless I live; yet 
      not I, but Christ liveth in me; and the life which I 
      now live in the flesh I live by the faith of the Son 
      of God, who loved me, and gave himself for me."       
                               Galatians 2:20 




From paul at prescod.net  Mon Aug 23 09:34:57 1999
From: paul at prescod.net (Paul Prescod)
Date: Mon, 23 Aug 1999 13:34:57 GMT
Subject: Man pages would be cool.
References: <37BC6986.A66B044E@starmedia.net>
								<199908191634.KAA01333@jhereg.perl.com>
								<14268.21198.934117.241668@anthem.cnri.reston.va.us>
								<14268.23007.239507.166209@weyr.cnri.reston.va.us>
						<y0j7lmq44wd.fsf@vier.idi.ntnu.no>
						<37BDD1B9.99588CB@prescod.net>
				<37BE24E0.79EA064E@montana.com>
				<37BE9EAB.CB7B1380@prescod.net> <37BEDBF3.BCA64E58@montana.com>
Message-ID: <37C14E01.3BB24034@prescod.net>

bowman wrote:
> 
> And try to use it like the Notepad type editor it resembles at first
> glance? I understand what you are saying, but I don't think chance
> encounters are very productive. 

That's how I've learned almost everything I know about every operating
system I've used. You realize you have something installed, figure it
must be useful and read the docs to figure out why. Isn't this principle
behind the very existence of menus at all?

 Paul Prescod



From mlh at idt.ntnu.no  Fri Aug 20 16:29:50 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 20 Aug 1999 22:29:50 +0200
Subject: Anyone know about lgrind?
Message-ID: <y0jyaf62o1t.fsf@vier.idi.ntnu.no>


Does anyone here know enough about how lgrind works to make it work
with Python? Has anyone done this?

Or... does anyone have some generic code for marking up Python? (I
remember Just's HTML-code... But I can't seem to find it. And besides
- it used the regex module <wink>)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From cmcurtin at interhack.net  Thu Aug 26 13:17:04 1999
From: cmcurtin at interhack.net (Matt Curtin)
Date: 26 Aug 1999 13:17:04 -0400
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl> <014101beecc0$e810e590$f29b12c2@secret.pythonware.com>
Message-ID: <xlx1zcqv4vj.fsf@gold.cis.ohio-state.edu>

>>>>> On Sun, 22 Aug 1999 17:07:14 GMT,
    "Fredrik Lundh" <fredrik at pythonware.com> said:

Fredrik> on the other hand, I'm not sure that would have helped.
Fredrik> judging from most threads that tchrist appears in (be it on
Fredrik> comp.lang.perl.misc, gnu.misc.discuss, or this group), things
Fredrik> often turn into a great flame fest when he's involved, by
Fredrik> some mysterious reason.

I haven't ever met Tom personally, but I've known him online for a
long time.

Tom complains loudly when something is broken or stupid.  Tom is very
direct; some would say "abrasive".  Tom is almost always right.  Most
people who are so taken aback by his presentation are unfamiliar with
him and/or are likely to take his arguments personally.

The lack of Python man pages is a problem for Unix folks.  We have an
online manual for a reason.  We're used to tools like `man' and
`apropos' that use the manual and help us find the things we seek.
Tom isn't "attacking" Python for not having it; he's (loudly) pointing
out a very legitimate gripe.  He similarly chastises the authors of
other programs (mostly from the Land o' Linux) that lack man pages.
Tom knows what sort of effort goes into the production of
documentation; he's put in a lot of time to make Perl's documentation
enviable.  He's done it for the good of the community, without
expectation of monetary compensation.  (As it turns out, his writings
have brought in some money, but I'm sure that if you ask, you'll find
that this wasn't the motivating factor.)  His gripe about the lack of
documentation (man pages) could fan a flame if someone chose to take
offense instead of fixing the problem by converting the documentation
into man format.

Tom has legitimate gripes against the GNU folks for claiming that his
excellent documentation isn't free.  With all due respect to RMS, one
of the two at the center of the "Perl docs aren't free" debate is a
linguist, and RMS isn't he.  RMS is certainly a master hacker and a
competent writer, but I have a strong distaste for his insistence that
only that which is perpetually available for review, change, and
redistribution qualifies as "free".  Tom's colorful presentation of
his side of this debate could trigger many a flamefest.

At the same time, Tom listens to legitimate gripes against his work.
He once yelled at a newbie for doing something stupid with his
headers.  I posted a response that was almost word for word that
pointed out a serious problem in his headers.  He didn't complain
about me, he didn't tell me to get a life, he didn't go sulk in the
corner.  He just fixed the problem and went on his merry way.

In my opinion, you'd be much better off learning what you can from Tom
instead of trying to criticize him for his perhaps suboptimal manner
of presentation and attempting to justify your own position.

Tom is trying to get the Perl community to learn Python and learn from
it.  Python has done some things well.  It has done some other things
not so well.  We, the users and designers of languages, have a great
deal to learn from each other.  Perl folks generally seem to recognize
that; it's sad that so many other language communities don't
reciprocate.  It's especially sad to see that Python's is one that
seems to like its glass house.

I hope that in the long run, the Python community as a whole will
prove me wrong.

-- 
Matt Curtin cmcurtin at interhack.net http://www.interhack.net/people/cmcurtin/



From gmcm at hypernet.com  Wed Aug  4 21:08:43 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Wed, 4 Aug 1999 20:08:43 -0500
Subject: Problems with sockets on Win NT
In-Reply-To: <7oaibf$763$1@nnrp1.deja.com>
Message-ID: <1278310965-17570891@hypernet.com>

Murali wrote:
> 
> I'm trying to exchange datagram (UDP) packets
> between a unix machine and a Win NT machine.
> 
> The UDP packets sent from Win NT machine are
> received properly on unix machine. However UDP
> packets sent from unix machine is not at all
> received on WinNT machine.
> 
> I'm using recvfrom method to receive the packets.

You sure? That's not what your code says:

>         test = my_sock.recv(1024)

- Gordon



From greg.ewing at compaq.com  Thu Aug 19 16:59:00 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Fri, 20 Aug 1999 08:59:00 +1200
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk> <19990816205639474.AAA315.413@max41127.izone.net.au> <lDvYnDAz1Hu3EwB6@jessikat.demon.co.uk> <37BB2E1D.C0733F85@boeing.com> <7pgd1g$hj6$1@newshost.accu.uu.nl>
Message-ID: <37BC7014.33B6FFF6@compaq.com>

Martijn Faassen wrote:
> 
> Funny, I read that 'the way g-d indented'. :)

Hey, what a great signature that would make!

Python: Programming the way g-d indented.
Greg



From robin at jessikat.demon.co.uk  Mon Aug 30 14:02:58 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 30 Aug 1999 19:02:58 +0100
Subject: Spread/Worksheet
References: <37CA8EAB.5DD1314@uniserve.com>
Message-ID: <uBBajJASdsy3EwRu@jessikat.demon.co.uk>

In article <37CA8EAB.5DD1314 at uniserve.com>, Justin Muir
<jkmuir at uniserve.com> writes
>Hello all.  I'm looking for a widget that has a spreadsheet(row,col)
>type scrollable interface.  Anyone seen anything?
>
>
>Justin.
>
tktable? the moodss thing? Lots of tcl.
-- 
Robin Becker



From mcfletch at vrtelecom.com  Tue Aug 24 10:58:41 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Tue, 24 Aug 1999 14:58:41 GMT
Subject: Newbie question about Re
In-Reply-To: <37C29893.D7B2C482@gnet.tn>
References: <37C29893.D7B2C482@gnet.tn>
Message-ID: <NDBBIAPHEKGOBIHCBLIICEFCCAAA.mcfletch@vrtelecom.com>

Well, here's my take on a "better way"... I don't see why you would want to
use the RE module when you aren't doing any pattern matching...

def splitlen( instring, desiredlength):
	temp = []
	current = 0
	lenstr = len( instring)
	while current < lenstr:
		newcurrent = current+desiredlength
		temp.append( instring[current:newcurrent] )
		current = newcurrent
	return temp

I haven't even tested this for correctness, but I think it's close to
correct.

Hope this helps,
Mike

-----Original Message-----
From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
Behalf Of MB
Sent: August 24, 1999 9:05 AM
To: python-list at cwi.nl
Subject: Newbie question about Re


Hi

I just want to split a string of n x (p characters) into a list of
p_length strings using the Re module
Example: Assume p=5 the string '12345ABCDE12345IJKLM' must be split in
['12345','ABCDE','12345','IJKLM'']
The way I  find is to make a replace before splitting:

p=re.compile('(.{5,5})',re.DOTALL)
m=p.sub(r'\g<0>###','12345ABCDE12345IJKLM')
# I use ### as separator
re.split('###',m)
['12345', 'ABCDE', '12345', 'IJKLM', '']
# And I have to remove the trailing ''

But I don't like this too much.
Is there a better way?

Thanks






From jstevens at basho.fc.hp.com  Fri Aug 13 23:03:12 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:03:12 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132302.RAA06674@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:02 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ce60@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :> Accessing an array and accessing a hash are different operations.
> :No, they aren't.
> :
> :That is the point of polymorphism.
> 
> Genuflect when you say that.   And realize that it's completely
> wrong-headed in these parts.

Wrong headed: as in, I don't worship at the same altar as you do?

John S.




From clgonsal at keeshah.penguinpowered.com  Fri Aug 27 03:23:42 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Fri, 27 Aug 1999 07:23:42 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <slrn7sasvb.397.kzaragoza@kzaragoza.ne.mediaone.net>
Message-ID: <slrn7scf7d.pa.clgonsal@keeshah.penguinpowered.com>

On Thu, 26 Aug 1999 17:07:43 GMT, Kris J. Zaragoza
<kzaragoza at mediaone.net> wrote:
> On Thu, 26 Aug 1999 14:33:40 GMT, Paul Prescod <paul at prescod.net> wrote:
> >I think it would be more pythonish if this would return a runtime (or
> >even compiletime) error:
> >
> >def foo():
> >	a=5
> >
> >b=foo()
> >
> >It shouldn't just return None. That's a source of errors.
> 
> The question here is, where is the error?
> ...
> Now when the call is made, you need to make a determination as to
> whether the calling line is in error, or the function itself is in
> error.  With your example, it could be either.  The author of the
> function may have erred in neglecting to insert a return statement at
> the end of the function.  If, on the other hand, the function does not 
> intend to return any values, then the author of the calling line made
> the error in simply assuming that it would.

The same could be said about many things that Python currently does
treat as errors. If I pass one parameter to a function that wants two,
is it my fault for leaving out a parameter, or the function's fault for
requiring it. Either way, Python assumes the caller is at fault. If I
wrote the function, and it's really the function that's at fault, I'll
realize that its definition should change, not the call.

> I tend to like Python's current way of handling things.  If a function
> does not explicitly return a value, None is returned instead.  It's a
> clean solution to an otherwise intractable problem.

It isn't intractable, unless you're trying to figure it out at compile
time.

> How do you propose fixing this in Python 2?  Would you enforce the
> separation of blocks that do and don't return values?  Would you
> include some sort of pragmas that tell the compiler and runtime that a
> value is not to be returned?  Other ideas?  (Personally, I wouldn't
> mind if this stayed as is.  I just read the documentation or doc
> string to see if a given function is meant to return a value.)

I frequently run into this same problem. Something I often do is write
functions that build up a result, and then I forget to return it. ie:

# Yes, I know there's a much easier way to do this. It's just for
# illustration...
def seqToString( seq ):
    result = '['
    if len(seq)>0:
        result = result + repr(seq[0])
        for item in seq[1:]:
            result = result + ', ' + repr(item)
    result = result + ']'


It's quite easy to go to all of the trouble of creating a result, and
then forget to return it. Later, when I call the above function, I get
back "None". If it instead raised an exception when I called it and
attempted to get a result from it, then these errors would be much
easier to catch.

Personally, I think falling out of functions and expecting None to be
implicitly returned is a very Perl-ish thing to do... Why don't we add
$_ to Python as well? I don't think it's too hard to explicitly return
None when that's what you want to do, and I would think/hope that most
existing Python code does this already.

It is possible, but more complicated, to make this a compile time error.
This would essentially require a function/procedure distinction with
restrictions about how you used return in each one. There would still
have to be a runtime error for cases where someone tried to get a result
from a procedure, but cases like the above example could be caught at
compile-time. (The above, being a function, must return a value.
Procedures would have to defined in some other way, like using some new
keyword "proc" in place of "def")

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From fredp at multimania.com.nospam  Fri Aug 27 08:59:18 1999
From: fredp at multimania.com.nospam (Fred Pacquier)
Date: 27 Aug 1999 12:59:18 GMT
Subject: Why do we call python a scripting language?
References: <19990827112918.12652.rocketmail@web129.yahoomail.com>
Message-ID: <8E2F958DCPaCmAnRDLM@news.dial.oleane.com>

guy_oliver at yahoo.com (Guy Oliver) said :

>Even after explaining it to them, and telling them my success
>stories, the improvements in productivity, the portability, etc, it
>seems that python suffers from the same view that hurts other
>interpreted languages, specifically, that interpreted languages,
>aka scripting languages, are for throw-away scripts, nothing more.
>
>I'm wondering if there is anything that can be done to combat this?
> With education, it goes away, but I expect people to equate script
>with simple for a long time, so the quickest solution seems to be
>try to distance 'python' from 'script'.

You have something there. Maybe another way would be to showcase the 'real' 
applications and their strengths, preferably in ways people can relate and 
compare to similar 'traditional' apps. Unfortunately it seems (to me) that 
such desktop, user-oriented apps in Python are still scarce ; much of the 
industrial-strength (like BSCW or Zope) happens on the backend and is Web-
oriented. Impressive for the initiate, but abstract and hard to marvel over 
for the profane. The one app that 'did it' for me, curiously, is PySol. 
It's a game, but a game engine, and a powerful one. And it shuffles around 
a lot of graphics (pun intended), which may surprise people biased by the 
'scripting language' tag. Right now I'm dabbling with similar stuff in 
wxPython instead of Tk, and I'm even more impressed.

Yet another selling point might be if IDLE or PythonWin or some other seed 
came to evolve into a full-blown IDE up to par with Microsoft's or 
Borland's, that might impress the developer population. After all, VC++ is 
done in MFC/C++ and Delphi in ObjectPascal, right ? But I seem to recall 
this was all discussed to death some months ago... :-)

-- 
YAFAP : http://www.multimania.com/fredp/



From kistler at fnmail.com  Sun Aug 22 07:07:35 1999
From: kistler at fnmail.com (Per Kistler)
Date: Sun, 22 Aug 1999 13:07:35 +0200
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk>
Message-ID: <37BFD9F7.7552BFB3@fnmail.com>

Phil Hunt wrote:
> Would anyone like to have a look at it, and possibly
> suggest some improvements? If so, tell me, and I'll put it on
> my website.

Let's give it a look:-) Where is it?
-Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From robin at jessikat.demon.co.uk  Mon Aug 30 10:59:38 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 30 Aug 1999 15:59:38 +0100
Subject: pop & append
Message-ID: <AEjBCBAaxpy3EwFp@jessikat.demon.co.uk>

Someone from the Zope list tells me that the list pop and append
operations are thread safe. Is this the case? How can one tell if
particular bits of python code are 'atomic'? 
-- 
Robin Becker



From cjc26 at nospam.cornell.edu  Mon Aug 30 23:39:24 1999
From: cjc26 at nospam.cornell.edu (Cliff, or a close facsimile)
Date: 31 Aug 1999 03:39:24 GMT
Subject: running python on a floppy?
References: <z6ex3.1541$1i4.128395@typhoon.mbnet.mb.ca> <37CA50EC.4FC03560@home.com>
Message-ID: <slrn7sm5m2.k0.cjc26@tankgrrl.bridget.mindriot.net>

Pada Mon, 30 Aug 1999 09:32:42 GMT, Doug Hellmann bilang:
| Why are people still trying to fit distributions like this on a
| *floppy*?  Why not use a Zip disk?  Granted, they are more expensive,
| but there is *so* much more room.

After losing two zip disks within a month of each other to the Zip Click
of Death, I vowed never to go near those things again. :P


-- 
cliff crawford   http://www.people.cornell.edu/pages/cjc26/
            There are more stars in the sky than there are
-><-        grains of sand on all the beaches of the world.



From m_krishnam at trillium.com  Thu Aug  5 12:45:54 1999
From: m_krishnam at trillium.com (m_krishnam at trillium.com)
Date: Thu, 05 Aug 1999 16:45:54 GMT
Subject: Problems with sockets on Win NT
References: <1278310965-17570891@hypernet.com>
Message-ID: <7ocf3u$i6e$1@nnrp1.deja.com>

In article <1278310965-17570891 at hypernet.com>,
  gmcm at hypernet.com wrote:
> > I'm using recvfrom method to receive the packets.
>
> You sure? That's not what your code says:
>
> >         test = my_sock.recv(1024)

Yes, there is a mismatch between code and description. I got it solved.
Thanks for the reply.

- Murali


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From kaz at ashi.FootPrints.net  Sun Aug 15 16:15:44 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Sun, 15 Aug 1999 20:15:44 GMT
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net>
Message-ID: <slrn7re808.77t.kaz@ashi.FootPrints.net>

On 15 Aug 1999 03:14:48 GMT, Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
>[strange programming]
>The most confidence uninspiring piece of code I ever had the displeasure to
>maintain contained something like this (allegedly C++, but mainly it's 
>intersection with C):
>
>	i = 0;
>	while( i < max_index ) {
>		/* some copying or update (not modifying i) goes here */
>		i = i + 1;
>	}
>
>This was written by a *very* expensive consultant working for a large
>multinational accountancy firm.

So what's wrong with that? Granted, the for loop construct would have
been better for the task instead of while, and the expensive consultant should
have seen that. But what about the i = i + 1 increment? It smacks of BASIC, but
there is nothing wrong with it.  It's not any less efficient than i++ or ++i.

Maybe the consultant is anally retentive about not using operators other than
assignment that have the side effect of modifying an object, and not using more
than one side effect in a full expression. That approach leads to uninsipiring
code, but nevertheless code that won't possibly have undefined behaviors due to
ambiguities.

In other words, with his ultra-conservative approach, the weenie is unlikely
to ever things like a[i] = i++.

I'm afraid you don't have a case, if the program otherwise fulfilled its
functional, performance and reliability requirements.

The guy programs for accounting firms. So it's a given that he can't possibly
be a great hacker who writes slick code. He does things that some of us
wouldn't touch with a ten foot pole.  His high cost could have something to do
with his domain specialization, and with a high profit margin on the part of
the contracting company (assuming he's not freelance).



From pointal at lure.u-psud.fr  Fri Aug  6 08:57:01 1999
From: pointal at lure.u-psud.fr (Laurent POINTAL)
Date: Fri, 06 Aug 1999 12:57:01 GMT
Subject: ANN: PIL 1.0 final
References: <008e01bedfec$7ffe7110$f29b12c2@secret.pythonware.com>
Message-ID: <37aadb42.108851580@news.u-psud.fr>

On Fri, 6 Aug 1999 09:17:24 GMT, "PythonWare" <info at pythonware.com>
wrote:

>: The Python Imaging Library (PIL) adds image processing capabilities
>: to your Python interpreter. This library supports many file formats,
>: and provides powerful image processing and graphics capabilities.
>: 
>: PIL 1.0 final is now available from:
>: 
>:     http://www.pythonware.com/

Yes, but the PIL link on this page gives me:
	http://www.pythonware.com/products/pil/
<<<
Forbidden

You don't have permission to access this URL on this server. The
reason for this might be some of the following:

  It has been locked out due to regulation violation (one or more
rules for Algonet has been ignored).
  It's locked out by the webmaster by some other reason.
  You have submitted the wrong URL (the correct URL to Algonet is
http://www.algonet.se).
  The administrator/owner of the file has locked it out for some
reason.

It might also be wrong permissions on the file you're trying to reach.
The administrator of the page and/or filename should issue the
command:

chmod 755 filename.html

...to make it world readable.

If you have any questions, please leave a message to me.

/Algonet Webmaster
>>>

A+

Laurent.




From jturner at ug.cs.su.oz.au  Mon Aug 30 01:34:59 1999
From: jturner at ug.cs.su.oz.au (Jeff Turner)
Date: Mon, 30 Aug 1999 15:34:59 +1000
Subject: Question: lists in classes
Message-ID: <37CA1803.2FA9261B@ug.cs.su.oz.au>

Apologies if this has been answered before...

class foo:
    list=[]

a=foo()
b=foo()
a.list.append(123)
print b.list

I would expect the above code to return "[]", since b has not been
modified since instantiation. However it returns [123].

How would I tell python that I only want to _define_ the class, not
instantiate it?

Thanks,
Jeff






From smithy at promail.pl  Sat Aug 21 14:31:19 1999
From: smithy at promail.pl (£ukasz Kowalczyk)
Date: 21 Aug 1999 18:31:19 GMT
Subject: Profiler - counting individual lines' exec time
Message-ID: <slrn7rtr32.hk.smithy@hoohoo.org>

Hi, everyone, it's my first posting on c.l.py.

I'm having hard time working with the Python profiler (profile.py). I have a
loop in my program which loops hundreds of thousands times and I want to know
which instructions in it take the most time to execute. Is that possible?

I imagine it could be done by tagging specific lines which execution time
I want measured but how do I do it? Using time.time() seems too messy and
not precise enough to me.

Moreover, documentation (html) provided seems to be incompatible with
reality :-) The Stats class does not provide any of the methods mentioned.

-- 
?ukasz Kowalczyk
http://tempac.fuw.edu.pl/~lukow/



From skip at mojam.com  Tue Aug 24 16:38:04 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 24 Aug 1999 15:38:04 -0500 (CDT)
Subject: Python equivilent of pack()?
In-Reply-To: <ytt6725sz5u.fsf@gilgamesh.cse.ucsc.edu>
References: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com>
	<14274.61530.251401.762453@weyr.cnri.reston.va.us>
	<ytt6725sz5u.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <14275.479.707949.774041@dolphin.mojam.com>

    Ben> I think it must not be too uncommon for folks to miss out on the
    Ben> fact that there's a function index to the library
    Ben> reference. Perhaps there should be a big fat bold link to it at
    Ben> both the top and bottom of the page.

    Ben> It would also be a good idea to use better icons than 'toc' 'm' and
    Ben> 'i' at the top of the page; I know I completely overlooked them
    Ben> until just now because they weren't very noticeable.

In fact, the Module Index is much more useful than the contents of the lib
ref manual once you're fairly well settled in.  I used to generate an index
myself before Fred started doing it (it was a pain to keep up with the TOC
format changes!).  Now I simply link to

    http://www.python.org/doc/lib/modindex.html

from my private home page and skip the lib manual's front page entirely.
Most everything I need is no more than two or three clicks away.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From simngolda at voyager.net  Sun Aug  8 23:57:30 1999
From: simngolda at voyager.net (Sim & Golda Zacks)
Date: Sun, 08 Aug 1999 23:57:30 -0400
Subject: Adding href and img to HTMLFMT
Message-ID: <7oljd7$1ni@newsops.execpc.com>

I am using the HTMLFMT module that came with the INTERNET programming
with Python and I wanted to add support for links and images but I am
having a problem. I can't figure out whether to add it as a descendent
of HTMLElement or of a Seq. The problem is it should be a Sequence
because There are no properties to it, but the URL is part of the tag
and I seem to be having a mental block when I try thinking about it. If
someone could help me with this, I would greatly appreciate it.

Sim




From possum at n-space.org  Thu Aug 12 17:17:02 1999
From: possum at n-space.org (Erik Stewart)
Date: Thu, 12 Aug 1999 21:17:02 GMT
Subject: Help!  $PYTHONPATH trouble
Message-ID: <37B33980.A5318181@n-space.org>

The variable $PYTHONPATH doesn't seem to have any effect on my system (RH
Linux 6.0, Python 1.5.1).  Here's what I get after defining $PYTHONPATH:

$ PYTHONPATH="home/possum/test/kwp"
$ echo $PYTHONPATH
home/possum/test/kwp
$ python
Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC egcs-2.91.66 19990314/Li
on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import sys
>>> sys.path
['', '/usr/lib/python1.5/', '/usr/lib/python1.5/test',
'/usr/lib/python1.5/plat-linux-i386', '/usr/lib/python1.5/lib-tk',
'/usr/lib/python1.5/lib-dynload', '/usr/lib/python1.5/site-packages']
>>> 

Shouldn't sys.path contain the value "home/possum/test/kwp"?



From quinn at dinar.ugcs.caltech.edu  Fri Aug 27 15:06:57 1999
From: quinn at dinar.ugcs.caltech.edu (Quinn Dunkan)
Date: 27 Aug 1999 19:06:57 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <37C5683D.91D61FD0@rubic.com>
Message-ID: <slrn7sdoel.821.quinn@dinar.ugcs.caltech.edu>

On Thu, 26 Aug 1999 16:15:57 GMT, Jeff Bauer <jbauer at rubic.com> wrote:
>Arguably, it may not be the best coding style, but I've 
>gotten used to functions returning None, if not otherwise
>specified.  If your proposal could be guaranteed to 
>generate a compile-time error (could it?), it might
>encounter less resistance, but I still wouldn't be
>very enthusiastic about the idea.  I can appreciate
>the issue you're addressing, though.
>
>Paul Prescod wrote:
>> 
>> I think it would be more pythonish if this would return a runtime 
>> (or even compiletime) error:
>> 
>> def foo():
>>         a=5
>> 
>> b=foo()
>> 
>> It shouldn't just return None. That's a source of errors.
>> This is obviously an incompatible change so I propose it 
>> for Python 2.

I agree with that Jeff that implicitly returning None is too handy, but I
think this could be solved by putting it in PyLint and having PyLidnt more
integrated with the distribution, or building some sort of compile/runtime
warning system into python (like gcc -Wall or perl -w) that can be arbitrarily
verbose.

Perhaps it would be possible to have python send all -Wxxx switches to
something like PyLint before compiling.  We already have -t, this would just
be extending that idea further.



From philh at vision25.demon.co.uk  Sat Aug 21 05:16:28 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sat, 21 Aug 99 09:16:28 GMT
Subject: Book -- or website -- about Tk & tkinter
References: <935172552snz@vision25.demon.co.uk> <37BE30BA.E45BB293@montana.com>
Message-ID: <935226988snz@vision25.demon.co.uk>

In article <37BE30BA.E45BB293 at montana.com> bowman at montana.com "bowman" writes:
> Phil Hunt wrote:
> > 
> > Can anyone recommend a good book or website that explains how Tk does
> > its geometry management (the grid, pack etc managers)?
> 
> Do you mean the low level nuts and bolts, or just how to use the
> managers effectively?

I want to know enough to write a Python/tkinter backend for my
Parrot GUI builder. So I'd guess I'd need to know some of the nuts 
and bolts.

> Many of the concepts are similar to Java, so any of the Java GUI
> tutorials might be useful,

Ah yes, Java's GridBagLayout widget. It was this, and IBM's
Visual Age visual composition tool, that provided the main 
inspiration for Parrot -- on  the grounds that ``there must be
a better way than this''.

> As far as tkinter, read the life preserver, Fredrik Lundh's in progress
> work, and the code in the Demo and Tools packages. Mostly, I'm afraid,
> read between the lines, and experiment.  

Is the Lif Preserver still useful? Lundh's _An Introduction to
Tkinter_ says it's a replacement for it.

But you're right -- experimentation will no doubt be necessary.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From abrahams at mediaone.net  Sun Aug 29 14:00:27 1999
From: abrahams at mediaone.net (Dave Abrahams)
Date: Sun, 29 Aug 1999 14:00:27 -0400
Subject: Porting Python
Message-ID: <Qtey3.430$LZ.46228@ndnws01.ne.mediaone.net>

Suppose I wanted to build a version of python for a platform that's not yet
supported. Does anyone have concise info on where the OS-dependencies are?
I'd like to get a sense of the scope of the job.

Thanks,
Dave



From che at debian.org  Mon Aug  9 21:52:32 1999
From: che at debian.org (Ben Gertzfield)
Date: 09 Aug 1999 18:52:32 -0700
Subject: Packing and unpacking IP addresses into 32-bit integers
References: <Pine.A32.3.90.990809163915.17213C-100000@elvis.med.Virginia.EDU> <NDBBIAPHEKGOBIHCBLIICEDKCAAA.mcfletch@vrtelecom.com>
Message-ID: <ytt907ks8pr.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Mike" == Mike Fletcher <mcfletch at vrtelecom.com> writes:

    Mike> Wouldn't it be something like: '>bbbb' or '!bbbb'

    Mike> instead (for portability to non-big-endian machines (I'm
    Mike> assuming the packed representation is supposed to have
    Mike> network byte order))?

Unfortunately Python isn't smart enough to assume that:

struct.pack('!bbbb', a, b, c, d)

means

struct.pack('!bbbb', d, c, b, a)

but it's easy enough to just change the order of the arguments, which
is what I did, since it's well-defined which order the arguments
will come in (they'll always come in a b c d and go out d c b a).

-- 
Brought to you by the letters N and G and the number 7.
"You have my pills!"
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From befletch at my-deja.com  Mon Aug  9 02:29:23 1999
From: befletch at my-deja.com (Bruce Fletcher)
Date: Mon, 09 Aug 1999 06:29:23 GMT
Subject: Parsing HTML tags with 're'
References: <7okkte$25a$1@nnrp1.deja.com> <7oknev$kac@dfw-ixnews15.ix.netcom.com>
Message-ID: <7olsfu$s4d$1@nnrp1.deja.com>

In article <7oknev$kac at dfw-ixnews15.ix.netcom.com>,
  aahz at netcom.com (Aahz Maruch) wrote:
> In article <7okkte$25a$1 at nnrp1.deja.com>,
> Bruce Fletcher  <befletch at my-deja.com> wrote:
> >
> >As I said before, I would also be interested to know if there is a much
> >better way to do this.  Did I miss a standard HTML parsing Python module?
> >If this approach is reasonable, feel free to scoop the code for your own
> >nefarious purposes.
>
> sgmllib/htmllib
> --
>                       --- Aahz (@netcom.com)

Hmmm.  Yes, I forgot about htmllib.  Some time ago, when I started
looking into this, I looked at the docs for htmllib and decided it wasn't
working for me.  Now sgmllib, that looks like maybe what I want.

Thanks,
- Bruce


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From news at dorb.com  Mon Aug 16 11:02:58 1999
From: news at dorb.com (Darrell)
Date: Mon, 16 Aug 1999 11:02:58 -0400
Subject: Difficulty configuring web server for Python
References: <7p9855$5tg$1@nnrp1.deja.com>
Message-ID: <JKVt3.1539$8P.41192@typhoon.nyroc.rr.com>

Do you think "home" has the same meaning for the server ?
#!/usr/home/campusch/usr/local/bin/python

--
--Darrell
<mbf2y at my-deja.com> wrote in message news:7p9855$5tg$1 at nnrp1.deja.com...
> My situation: I just obtained a "virtual web server" account at a web
> hosting place, where I have full access to UNIX on the box and can
> modify my apache config files, etc.  They advertise that they support
> Perl, C, and C++ for CGI scripting.  Needless to say, first thing I did
> when I got the account was to build a copy of Python.  Python appears to
> work fine.
>
> I built a simple CGI script and kept getting error 500, so I simplified
> to the below script:
>
> #!/usr/home/campusch/usr/local/bin/python
>
>
> print "Content-type: text/html"
> print
> print "<HTML><BODY><H1>Hello World!</H1></BODY></HTML>"
> print
>
>
> 1) Python is installed in /usr/home/..../python
> 2) running "python simp.py" from the command prompt gives me the
> expected result.
> 3) running "./simp.py" from that directory runs the program just fine.
> 4) python itself has been chmod ugo+x so the web server process should
> be able to run it.
> 5) This script has been chmod ugo+x so the web server process should be
> able to run it.
> 6) The script is in the correct directory; I installed a "supported"
> perl script in this directory and got it to work no problem.
>
> This is what gets written to the server log when I try to run my script:
>
> [Sun Aug 15 21:59:31 1999] [error] (2)No such file or directory: exec of
> /usr/local/etc/httpd/cgi-bin/simp.py failed
> [Sun Aug 15 21:59:31 1999] [error] Premature end of script headers:
> /usr/local/etc/httpd/cgi-bin/simp.py
>
> While the "no such file or directory" threw me, this is not a file not
> found error... when I tried to run a script that really wasn't there, I
> got:
>
> [Sun Aug 15 22:10:18 1999] [error] script not found or unable to stat:
> /usr/local/etc/httpd/cgi-bin/foo.py
>
> I noticed in my srm.conf file that they had the line
> AddType application/x-httpd-cgi .cgi
>
> in there, so I renamed the .py file to .cgi.  No help at all.
>
> As for the OS I'm running, here's what python prints on the first line
> when I run it:
>
> Python 1.5.2 (#3, Aug 15 1999, 19:04:09)  [GCC 2.7.2.1] on bsdos3
>
> My first recourse (after spending 3 hours trying to chase this down) was
> to contact tech support; they agreed with me that Python is a robust
> scripting language, but in order to make sure that they have time to
> handle all their clients' requests for supported software, they have had
> to tighten their policy and they aren't allowed to even help me a teensy
> bit when I'm using unsupported software.
>
> Frustrating as heck.  Anyway, if anyone here has any suggestions as what
> to try next, I would appreciate it.  I can't recompile apache but I can
> change the config files.  I can however run a script to find out the
> version of Apache I'm using, and here it is:
>
> Server version: Apache/1.3.1.1 SSL/1.15 PHP/3.0.3
> Server built:   Jun  3 1999 11:15:00
> Standalone Configuration
>
> I don't want to have to learn perl and I really don't want to do my
> scripting in C++.  So if anyone here has any suggestions, I'd greatly
> appreciate it.
>
> Thanks,
> -Fred
>
>
> --------
> Michael "Fred" Fredericks
> PhD Student, Computer Science
> University of Maryland College Park
> fred.at.cs.dot.umd.dot.edu  ---- note I never check my dejanews mail.
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.





From ron at library.ucsf.edu  Fri Aug 27 19:18:08 1999
From: ron at library.ucsf.edu (Ron Klatchko)
Date: Fri, 27 Aug 1999 16:18:08 -0700
Subject: Python article on Builder.com
References: <slrn7rpqbj.atl.kjz@kzaragoza.ne.mediaone.net> <MPG.122f52adbcf97bbd989698@news.nettally.com>
Message-ID: <37C71CB0.A3851A2D@library.ucsf.edu>

Tom Funk wrote:
>  - the value of Object Orientation in a scripting language

OO is another methodology for building software.  Some classes of
problems are best addressed with OO.  When you are faced with such a
problem, OO is indispensable.

Which reminds me of one of my favorite quotes.  It's from 1991 when OO
had become very popular:

... You have to think.  There is no substitute for thinking.  There are
people who think that "object-oriented" will solve your problems for
you.  Rubbish!  "Object-oriented" is (only) a way for you to attack your
problems.  It is not a magic wand that makes the problems go away
without thought or effort.  --Bjarne Stroustrup

moo
----------------------------------------------------------------------
          Ron Klatchko - Manager, Advanced Technology Group           
           UCSF Library and Center for Knowledge Management           
                        ron at library.ucsf.edu



From fredrik at pythonware.com  Sun Aug 22 13:07:14 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Sun, 22 Aug 1999 17:07:14 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com> <7pniv5$mlm$2@newshost.accu.uu.nl>
Message-ID: <014101beecc0$e810e590$f29b12c2@secret.pythonware.com>

Martijn Faassen <m.faassen at vet.uu.nl> wrote:
> Oh, I don't think he's that bad. He *does* flame back when he feels people
> are flaming him, and sometimes his finger is a bit too near the flame
> trigger, but he's nice and reasonable to the people who are nice and 
> reasonable to him.
> 
> So let's try to be nice and reasonable to him?

you mean like saying "you're right, python is completely un-
documented" and "python 1.4 is still the current version" and
"you're right, the assignment statement knows how to copy
some data types but not all", etc?  that's fine with me.  the
customer is always right, you know ;-)

on the other hand, I'm not sure that would have helped.
judging from most threads that tchrist appears in (be it
on comp.lang.perl.misc, gnu.misc.discuss, or this group),
things often turn into a great flame fest when he's in-
volved, by some mysterious reason.

maybe we should add a note to the newsgroup FAQ?

</F>




From wware-nospam at world.std.com  Fri Aug  6 16:03:43 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Fri, 6 Aug 1999 20:03:43 GMT
Subject: Python microthreads
References: <000301bedef5$89096280$662d2399@tim>
Message-ID: <FG27q7.JEr@world.std.com>

Tim Peters (tim_one at email.msn.com) wrote:
> ...you went out of your way to change ceval.c to use both
> brace and indentation styles Guido hates <0.9 wink>.

For my own private tinkering, I usually use the default indentation
modes offered by "indent" and Emacs C-mode. This is a matter of
laziness, not principle. If it ever mattered, I'd happily defer to
another style.

> Python already bytecode-slices "real threads" every N
> opcodes (set via sys.setcheckinterval)

That looks like the right way to do it. Thanks for pointing
that out.

> ...fake threads of any flavor [would, without enormous effort] all
> block at I/O calls.

I envision using microthreads for dumb little agents in swarm-like
simulations (game examples: robot tanks, PacMan monsters) where
nothing like file I/O would ever come up. But yeah, it would be good
if there were eventually some documentation with discussion of
limitations.

For those who want to follow along at home: A PyThreadState has two
relevant fields. tstate->ticker counts down executed bytecodes
starting from tstate->interp->checkinterval, the same thing my
num_steps counter is doing, except that all the machinery already
exists in the core release. sys_setcheckinterval(), defined in
sysmodule.c, sets tstate->interp->checkinterval. "ticker" is a field
of a PyThreadState, "checkinterval" is a field of a
PyInterpreterState, both defined in Include/pystate.h.

The use of "ticker" as a bytecode-slicing mechanism is enabled by a
static flag called "things_to_do" in ceval.c. My "uthread.step"
function would need to set things_to_do, call eval_code2_loop, and
reset things_to_do to its prior state. I'm not yet sure how to handle
threads whose status is "still working". Maybe I need to add a
WHY_STILLWORKING enum value, and possibly flag the thread or frame in
some way to represent that status.
-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From gmcm at hypernet.com  Sat Aug 28 11:05:54 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 28 Aug 1999 10:05:54 -0500
Subject: Python Type System: An idea for unification
In-Reply-To: <37c72e93.1411973891@news.triode.net.au>
Message-ID: <1276273608-41434292@hypernet.com>

John (Max) Skaller wrote:

 [Stuff interesting enough to perhaps revive the Types SIG...]

 [Briefly, and subject to my misinterpretation, it appears that
  Viper objects always have a type obj; said type obj may be
  a class instance; said type obj is appended to the normal
  attribute search path]

>  class PyIntType:
>   def succ(x): return x + 1
>   def pred(x): return x - 1
>
> To use this we try out the Viper code:
> 
>  x = 1
>  print x . succ()

At first glance, these are class methods. But not really, because the 
bound method object binds the explicit arg 'x' to the user object (in 
this case also named 'x'). But that's not

 class PyIntType:
  def succ(self): return self + 1
  ...

because "self" in this case would be the instance of PyIntType, not 
the user object (which "has a" instance of PyIntType). So _something_ 
is special here. Ah, you are actually re-binding (the bound to 
type(x) method) "succ" to the user object ("1"). Right?

 [Generic programming...]

>  class ListOf:
>   def __init__(self,etype):
>    self.etype = etype
>   def __call__(self): 
>    obj = new_object(self)
>    obj.data = []
>    return obj
>   # ...
>
> Here, we use a class ListOf as a _generic_ type for all
> lists of the same kind of type, and we use an _instance_
> of that class for the type: the argument to the __init__ method
> specifies the type:
> 
>  ListOfInt = ListOf(IntType)

So ListOfInt is an instance of ListOf. Unbound methods defined in 
ListOf are bound methods in ListOfInt. Presumably, they will be 
rebound to the object using ListOfInt as its type object.

> will construct the type object for a list of ints.  And here is an
> actual list of ints:
> 
>  x = ListOfInt()
>
> Here, the __call__ method defined for the generic type ListOf
> is called, because ListOfInt is an instance of the class ListOf. The
> 'new_object' function creates a new object, and sets the type of the
> object to its argument, which is an instance of ListOf for which the
> 'etype' parameter is IntType.

Is "new_object" a builtin?

> We need to check where the binding occurs. In the
> lookup on ListOf, a function found there is callable,
> and so is bound to the instance ListOfInt.
> But we want such a method to be bound the instance x
> as well. So a generic method in ListOf needs TWO object
> arguments:
> 
>  1) the first argument binds to ListOfInt
>  2) the second argument binds to x

One of us is missing something here! There doesn't seem to be anyway 
to differentiate between:

 class ListOf:
  def method1(arg1, arg2): ...

being called from the user object x as
 x.method1()   #generic
  and/or
 x.method1(somearg) #non generic

It seems to me you either need an explicit way of marking them as 
different when writing ListOf, or (I think more generally), writers 
of type classes need to always handle 2 "self"s:

 class PyIntType:
   def succ(self, obj): return obj + 1
   # "self" is the instance of PyIntType,
   # "obj" is the "self" of the user object - some obj whose type
   # object is an instance of PyIntType   
   ...

Then, of course, you are not rebinding a method; you are binding the 
2nd arg of an already bound method.

Cool stuff! And you successfully avoided ever using the dreaded 
"meta" word...

- Gordon



From cgw at fnal.gov  Fri Aug  6 10:37:19 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Fri, 6 Aug 1999 14:37:19 GMT
Subject: Linux why not fully dynamic?
In-Reply-To: <aOAnwBA2rjq3EwnZ@jessikat.demon.co.uk>
References: <aOAnwBA2rjq3EwnZ@jessikat.demon.co.uk>
Message-ID: <14250.62239.580394.67647@buffalo.fnal.gov>

Robin Becker writes:
 > Why doesn't ./configure set up for a libpython.so.1.5.2 or whatever?

This is an FAQ, see:
 
 http://www.python.org/doc/FAQ.html#3.30




From david.stegbauer at cz.opel.com  Wed Aug  4 03:20:26 1999
From: david.stegbauer at cz.opel.com (david.stegbauer at cz.opel.com)
Date: Wed, 4 Aug 1999 07:20:26 +0000
Subject: A question of style
Message-ID: <C12567C3.002D97D7.00@derumg01.cyberlink.eds.com>


Roy Smith wrote:
> I'm thinking of giving my base widget class a cursor method, which looks like:
>
>def cursor(self):
>   return self.page.handler.server.db.cursor()
>
>and then I can just do self.cursor() inside my widgets.

If you make alias like
cursor = self.page.handler.server.db.cursor
then you can simply call cursor() function - it is much faster than your
version.

David





From loredo at spacenet.tn.cornell.edu  Thu Aug 19 14:12:44 1999
From: loredo at spacenet.tn.cornell.edu (Tom Loredo)
Date: Thu, 19 Aug 1999 14:12:44 -0400
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com>
		<14265.29899.470021.309355@dolphin.mojam.com>
		<14265.47748.241386.291144@buffalo.fnal.gov>
		<wkpv0mgna5.fsf@turangalila.harmonixmusic.com> <14265.58497.578663.958576@buffalo.fnal.gov>
Message-ID: <37BC491C.F6654145@spacenet.tn.cornell.edu>

Charles G Waldman wrote:
> 

> Correct.  However, I'm curious, what is the definition of a "perfect
> shuffle"?  

It's not a computer science article, and thus not completely relevant
to this thread, but if you want to learn more than you'll ever need
to know about the statistics of shuffling, Persi Diaconis is the man.
A standard reference is:

Trailing the Dovetail Shuffle to Its Lair",
by Dave Bayer & Persi Diaconis,  Annals App. Prob, Vol 2 # 2, May 1992, p 294.

This is the fellow who worked out the item reported in the popular press
a few years ago that it takes 7 shuffles to adequately shuffle a deck
of 52 cards.

Peace,
Tom Loredo



From tim_one at email.msn.com  Fri Aug 27 00:27:39 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 27 Aug 1999 00:27:39 -0400
Subject: Fuzzy string matching?
In-Reply-To: <37C5F53D.B201DD1B@easystreet.com>
Message-ID: <000a01bef044$7f70e4e0$592d153f@tim>

[Al Christians]
> I've gotten good results with ad hoc algorithms using a longest
> common contiguous substring routine. ...

Note that there's a capable & optimized SequenceMatcher class in

    Tools/Scripts/ndiff.py

that does exactly that, also accepting an optional characterization of "junk"
sequence elements, and with methods to pump out a list of "how to change
sequence1 into sequence2" edit operations, or just return a "similarity ratio"
(a float in [0.0, 1.0], from "nothing in common" to "identical").  Any kind of
sequence is OK as input, so long as the elements are hashable and support a
__cmp__ method that can distinguish equal from not-equal (it doesn't have to
make sense of < or >).

> ...
> How to combine these lengths into a scalar measure of match is
> the really ad-hoc part of it.

I defy you to prove that SequenceMatcher's combination is wrong <wink>.

flamingly y'rs  - tim





From varner at PENELOPE.CHEM.UIC.EDU  Wed Aug 18 14:06:02 1999
From: varner at PENELOPE.CHEM.UIC.EDU (varner at PENELOPE.CHEM.UIC.EDU)
Date: 18 Aug 1999 13:06:02 -0500
Subject: Password Prompt
Message-ID: <82iu6dyned.fsf@PENELOPE.CHEM.UIC.EDU>

Hi,

I'm writing a Tkinter interface to a program that prompts for a
password.  I would like to pop a dialog when the program sends the
prompt, but I'm having a little trouble figuring out how to do that.
Right now I'm doing

	message = os.popen (command, 'r').readlines ()

but the program sits at this line waiting for a response to the
password prompt.  Any hints?
-- 
Sam Varner
Department of Chemistry
University of Illinois at Chicago



From cbbrowne at news.hex.net  Mon Aug 30 22:18:06 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Tue, 31 Aug 1999 02:18:06 GMT
Subject: Scripting Language
References: <7qd6eq$iq1$1@news.is.co.za> <m3iu5xhexv.fsf@pcrm.win.tue.nl>
Message-ID: <yXGy3.427$Jl.19931@news6.giganews.com>

On 30 Aug 1999 10:04:44 +0200, Stephan Houben
<stephan at pcrm.win.tue.nl> wrote: 
>"Jacques Oosthuizen" <jacques at integra.co.za> writes:
>
>>We are looking at a multiplatform scripting language. So far python
>>looks the best , any comments.
>
>Well, it satisfies the whole list of requirements you posted. ;-)
>(But then again, so does INTERCAL.)
>
>Seriously, without some more context, this question is meaningless.
>
>P.S. Is Turing-completeness perhaps of any importance?

<asbestos>
Python isn't Turing-complete, right?  
</asbestos>

Python seems pretty good in many ways; without more precise
specifications of requirements, it is indeed difficult to assess the
preferability of any given language.

- Python appears pretty easy to embed, as well as easy to embed C code
  into.

- Perl is probably more popular, and has vast libraries of "modules,"
  possibly more than are available for Python, but the schemes for
  embedding compiled code into Perl and vice-versa seem somewhat more
  "daunting."

- TCL was designed with inter-embeddability in mind, so it might be a
  not-outrageous choice.

- There are Scheme implementations designed with embedding in mind too.
-- 
I have stopped reading Stephen King novels.  Now I just read C code
instead.
-- Richard A. O'Keefe
cbbrowne at ntlug.org- <http://www.hex.net/~cbbrowne/langscript.html>



From arcege at shore.net  Tue Aug 10 15:08:26 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 10 Aug 1999 19:08:26 GMT
Subject: posix.error is a tuple, not an object or a string as docs say
References: <yttzozzbm11.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <KM_r3.1361$EG4.234560@news.shore.net>

Ben Gertzfield <che at debian.org> wrote:
: Today I was trying to use the posix.fdopen() function (through
: os.fdopen() as recommended in the documentation) so that I could open
: a pre-existing file descriptor passed in through sys.argv to a normal
: Python file object.

: Obviously I want to be able to catch errors in case the file
: descriptor passed through sys.argv isn't legal, so I read the nice
: documentation for posix.error:

:    error
:           This exception is raised when a POSIX function returns a
:           POSIX-related error (e.g., not for illegal argument types). The
:           accompanying value is a pair containing the numeric error code
:           from errno and the corresponding string, as would be printed by
:           the C function perror(). See the module errno , which contains
:           names for the error codes defined by the underlying operating
:           system.

:           When exceptions are classes, this exception carries two
:           attributes, errno and strerror. The first holds the value of
:           the C errno variable, and the latter holds the corresponding
:           error message from strerror().
:           
:           When exceptions are strings, the string for the exception is
:           'os.error'; this reflects the more portable access to the
:           exception through the os module.

: This doesn't seem to be quite true, however:

: Python 1.5.1 (#1, Dec 17 1998, 20:58:15)  [GCC 2.7.2.3] on linux2
: Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
:>>> import os
:>>> try:
: ...     input = os.fdopen(42)
: ... except os.error, error:
: ...     print error 
: ... 
: (9, 'Bad file descriptor')

: 'error', in this case, is a tuple, not an object. Am I incorrect in
: thinking this? If so, how do I access the aforementioned attributes
: errno and strerror?

: Obviously it's no problem to just say:

: except os.error, error:
:     errnum, errstr = error

: but this contradicts the documentation for posix.error -- and there is
: no documentation for a separate os.error, so I assume they are one and
: the same.

: I guess I'm missing something major, but it's odd that the docs don't
: match reality at all.

: Ben

Hi Ben,

It sounds like you are using a release of either Python 1.5 or 1.5.1.
If you are using Python 1.4 or earlier, then yes, it is a tuple.  If
Python 1.5 or later, then it is an instance of an exception class.

And this is polymorphism at it's best.  From the appearance, it looks
like a tuple, but if you perform some interspection on the "error"
object, you will see it is an instance of IOError and inherits the
"__str__" method from the Exception class, which converts the arguments
to the constructor into a string.

The long and short of this is that you can use it like a tuple if more
than one argument was given, and as the object otherwise.

For some of your own reassurance, print the type of "error" in the
except clause.

  -Arcege




From cbbrowne at news.hex.net  Thu Aug 12 23:25:44 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Fri, 13 Aug 1999 03:25:44 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk> <luemh851ym.fsf@stest.ch>
Message-ID: <YeMs3.6620$gO1.175350@news2.giganews.com>

On 12 Aug 1999 19:43:45 +0200, Christian Egli <christian.egli at alcatel.ch> wrote:
>>>>>> "Phil" == Phil Hunt <philh at vision25.demon.co.uk> writes:
>>>>>> "Paul" == Paul Prescod <paul at prescod.net> writes:
>Paul> Glade is a GUI application that allows you to build a window
>Paul> visually.
> 
>Phil> Okay, but I saw a reference somewhere to Glade having a
>Phil> declarative file format that could be converted into code to
>Phil> build the GUI. I was mostly talking about the declarative file
>Phil> format.
>
>Maybe you meant the storage format of Glade which is XML which in
>turn can be used with Libglade to build the GUI dynamically.

Sounds Really A Lot like CORBA DII...

>Paul> Glade does that too, as long as your "favorite language" is
>Paul> C. :)
>
>Not only. See below
...
>AFAIK Libglade does exactly that.

Question: How do you tie events to widgets?

Is it, maybe, perchance, via pointers to C functions???

-- 
Rules of the Evil Overlord #65. " If one of my dungeon guards begins
expressing concern over the conditions in the beautiful princess'
cell, I will immediately transfer him to a less people-oriented
position." 
cbbrowne at hex.net- <http://www.ntlug.org/~cbbrowne/lsf.html>



From kernr at mail.ncifcrf.gov  Mon Aug 30 01:56:27 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Mon, 30 Aug 1999 05:56:27 GMT
Subject: But Sir, It's Wafer Thin!!!
References: <37CA141D.41285EFD@megsinet.net>
Message-ID: <37ca18a6.33544983@news.erols.com>

On Mon, 30 Aug 1999 01:18:23 -0400, David Oppenheimer
<davidopp at megsinet.net> wrote:

>I only have a small question.  In fact you might say its wafer thin.
>
>Q: What path declarations do I need to place into my autoexec.bat file
>so that my computer will be able to run Python Scripts?

[snip TCL stuff]

>I used the find command to
>locate both of these dll's and they are here C:\TclPlug\2.0\bin

Only there?

>I didn't do anything weird when installing Python on my Win98 machine.
>Chose the 1.5.2 install downloaded and installed it (I am under the
>understanding that this includes all needed Tcl software for Tkinter).

Did you let the TCL/TK installer run?  If so, you should have another
directory with the TCL DLLs.  That directory should probable be used
rather than the TclPlug directory.

>I also installed the additional 3MB win32 all.
>Python and Pythonwin are located here C:\Program Files\Python\Pythonwin

Where is "python.exe"?  In "C:\Program Files\Python" or 
"C:\Program Files\Python\Pythonwin"?  I'll assume the first for my
statements below.

>Do I need just one path declaration or multiple path declarations and
>how would they look?  

You should have a pre-existing line looking this:

set PATH=C:\WINDOWS;C:\WINDOWS\COMMAND;...

If not, just ignore it for now.  Perhaps someone more esoteric Windows
knowledge than I could comment (Tim?).

You should then have a second line that looks like this:

set PATH=%PATH%;C:\PROGRA~1\PYTHON;C:\TCL\BIN

C:\TCL\BIN should be whatever path that tcl80.dll and tk80.dll (from
the Python-provided installer) are in.

You might encounter problems if another TCL path is in the PATH
variable (say, if the TclPlug installation altered your PATH).  Then,
you should move the "C:\TCL\BIN" before "%PATH%".

>My only other experience with this kind of stuff
>was with setting up a pathway for a compiler and it was well documented
>that my path declaration for the Mingw32 compiler should look just like
>this....   PATH=c:\mingw32\bin;

The proper way to do that is

set PATH=%PATH%;c:\mingw32\bin

IIRC, "PATH=c:\mingw32\bin" will work the same way, but it is not
guaranteed to.

>Please help...I need to get over this little bump in the road and then I
>will be on my way to complete mastery of this language...lol  Just a
>little help...wafer thin...just point me in the right direction...tell
>me I failed to adequately search the docs...anything...except don't send
>the Spanish Inquisition.

HTH.

>David O.

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From fran at nospam.com  Thu Aug 26 17:11:03 1999
From: fran at nospam.com (HERBS)
Date: Thu, 26 Aug 1999 17:11:03 -0400
Subject: Python Newbie Seeks Help
Message-ID: <7q4amg$8ta$1@ash.prod.itd.earthlink.net>

I'm new to python and having a hard time digging up the info
i need on a couple of issues.

I've searched the "net" and looked extensivley through
python.org.

What i am looking to do is extract data from an old
non ODBC compliant informix database ( "version 4.x
of informix SE) through informix 4gl programs and pass the
data to a server witten in python that would the pass the data
to the clients that have requested it.

My two area's of concern with python are:

1) I would some form of a "real world" example of how to
     use SocketServer.

2) Any instruction on passing data to a python program
     would also be very helpful.


Any help is much appreciated.
Thanks







From skip at mojam.com  Mon Aug 16 12:37:16 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 16 Aug 1999 11:37:16 -0500 (CDT)
Subject: Problem with "open"...
In-Reply-To: <Pine.SGI.3.96.990816111849.37560A-100000@voyager.vrve.iupui.edu>
References: <Pine.SGI.3.96.990816111849.37560A-100000@voyager.vrve.iupui.edu>
Message-ID: <14264.15779.609468.676383@dolphin.mojam.com>

>>>>> "Michael" == Michael Boyles <mjboyles at voyager.vrve.iupui.edu> writes:

    Michael> ... and I now get an error from the following code:

    Michael> class effect:
    Michael>   ...
	
    Michael>   def render():
    Michael>     ...
    Michael>     self.createViewFile("0000.vf")
    Michael>     ...

    Michael>   def createViewFile(self, filename):
    Michael>     f = open(filename, "w")
    Michael>     f.write("rview")
    Michael>     f.close()
	
    Michael>   ...

	
    Michael> The error is on the "open" function. It reads:
	
    Michael>   Exception in Tkinter callback
    Michael>   Traceback (innermost last):
    Michael>   File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 726, in __call__
    Michael>     return apply(self.func, args)
    Michael>   File "/usr/people/mjboyles/effect/bin/effectGUI.py", line 156, in render
    Michael>     self.menuListener.render()
    Michael>   File "effect.py", line 423, in render
    Michael>     f = open(("%s.vf" % name), 'w')
    Michael>   TypeError: illegal argument type for built-in operation

    Michael> Any suggestions?

Two things come to mind.  First, the line quoted in the traceback is not the
line you posted as the open call for createViewFile, so it looks like you're
not looking in the correct place for the error.  Second, the error is
probably occurring while evaluating

    "%s.vf" % name

The variable name is not a string.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From gmcm at hypernet.com  Thu Aug 12 15:19:15 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Thu, 12 Aug 1999 14:19:15 -0500
Subject: Copying Huge Files ? (Newbie!)
In-Reply-To: <37b30e52.43353218@news.usenetserver.com>
Message-ID: <1277639779-21637942@hypernet.com>

Matthias Barmeier wrote:
> 
> 
> I want to copy some files across the disks of my server using python
> parse a script and then do what I want.
> 
> But while reading the manual I cannot find something like
> os.copyfile or os.movefile.
> 
> Are there some functions except open/read/write that can do the copy
> for me ???

shutil.py in lib/.

- Gordon



From ivanlan at callware.com  Wed Aug  4 08:00:55 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Wed, 4 Aug 1999 12:00:55 GMT
Subject: Get 'ifconfig' information via Python
References: <87yafs8ruu.fsf@linuxcare.com> <37A81400.7474FD89@lemburg.com>
Message-ID: <37A82B77.FB3D03E2@callware.com>

Hi All--

"M.-A. Lemburg" wrote:
> 
> David N. Welton wrote:
> >
> > So, I think I'm getting close...
> >
> > import fcntl
> > import IN
> > import struct
> >
> > s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
> >
> > res = fcntl.ioctl(s, IN.SIOCSIFHWADDR, ??????)
> >
> > I can't seem to find what goes in ?????.
> >

[snip]

> The .ioctl() function can take a string or an integer as
> third argument. 

[snip]

> 
> Building these structs can be done using the Python struct
> module... playing around with this can probably crash your system
> though.
> 
> In the end, I think you're better off hacking together a
> new extension module.
> 

Why on earth would reading this information into a struct crash his
system?  Or am I missing something here?  At worst, writing a messed-up
struct might hose his ethernet connection, but I really doubt that he
would be any more likely to crash his system using Python than C.

<c-is-your-friend-while-it-gnaws-your-hand-off>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From catfood at apk.net  Thu Aug 19 10:19:20 1999
From: catfood at apk.net (Mark W. Schumann)
Date: 19 Aug 1999 10:19:20 -0400
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu>
Message-ID: <7ph3p8$7ci@junior.apk.net>

In article <37b62939 at cs.colorado.edu>,
Tom Christiansen  <tchrist at mox.perl.com> wrote:
>Gosh yes.  I think it was the rogue source that had something like:
>
>    #define until(expr) while(!expr)
>    #define otherwise   break; default:
>
>I don't remember whether it also had these, but I have
>seen them elsewhere:
>
>    #define forever()  for(;;)
>    #define unless(expr) if(!expr)

I've seen

#define ARBEGIN {
#define AREND   }

in Clipper (a somewhat C-ish language in this regard) code.

Eh?




From tchrist at mox.perl.com  Thu Aug 19 18:57:54 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 15:57:54 -0700
Subject: More random python observations from a perl programmer
References: <199908191819.MAA04074@jhereg.perl.com> <37bc71cc.19700984@news.erols.com>
Message-ID: <37bc7de2@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    kernr at mail.ncifcrf.gov (Robert Kern) writes:
:This is a holdover from the time before the
:source and documentation were split into separate distributions.  

Why was this done?  Is this a good thing?   You can't do that with Perl
modules' documentation because all the module doc is embedded.   We did
this to avoid a sync problem, and so people could never not have the docs.
In fact, they can even be links to the same file, e.g. File::stat.3
linked to and File/stat.pm.  Or do you just mean core docs, not
module docs were split?  

--tom
-- 
"IMHO, CAPS LOCK should be somewhere more
convenient; e.g., in another building."  --jgreely



From thantos at brimstone.mecha  Sat Aug 28 19:28:00 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Sat, 28 Aug 1999 23:28:00 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <slrn7sasvb.397.kzaragoza@kzaragoza.ne.mediaone.net> <slrn7scf7d.pa.clgonsal@keeshah.penguinpowered.com> <m3wvuh8yo3.fsf@pcrm.win.tue.nl>  <slrn7sf2m1.1gv.clgonsal@keeshah.penguinpowered.com> <slrn7sgesi.lbd.thantos@brimstone.mecha>  <slrn7sgq7j.4ee.clgonsal@keeshah.penguinpowered.com>
Message-ID: <slrn7sgrul.sv3.thantos@brimstone.mecha>

On Sat, 28 Aug 1999 22:55:59 GMT, C.Laurence Gonsalves
<clgonsal at keeshah.penguinpowered.com> wrote:

>should return None though. They should return *nothing-at-all*. The fact
>that some value (None) is returned in these cases is a source of errors,
>because function writers sometimes forget to return a value, and
>function callers sometimes try to get a value from something that isn't
>intended to return anything.

Python is not a Bondage and Discipline language; with that nature
comes a very important freedom, the Freedom To Make Mistakes.  Yes,
you can write code that doesn't run or is broken in Python.  Oddly,
however, this doesn't differ from any of the other languages which are
heavily into B&D, its just that Those Others make you jump through
hoops and learn to tie ropes just for the opportunity to make
mistakes.

I think you're critically overestimating the enormity of this
'problem,' in fact I don't think its a problem at all.  The amount of
flexibility that comes as a result for various coding styles seriously
overshadows the fact that, yes, someone can shoot themselves in the
foot with it.  By that measure, we should remove classes from the
language since they can be a source of errors.  Reducido ad absurdiam.

>There are several ways to have the language reduce these kinds of
>errors. I've suggested at least three possibilities. Some of these

Reducing these kinds of 'errors' isn't the function of a language.  If
a language construct can be misused, its not the fault of the language
design, its the fault of a programmer.  Rather than misplace the
effort of 'fixing' the problem of Guido's head, I feel, like several
others posting to this thread, that its far more a cause to put in
another check for PyLint.  If the programmer chooses not to PyLint his
code, that's his choice.  If he chooses to use a construct that can be
dangerous, that's his choice, too.

I have only encountered this kind of error a few times in the years
I've been programming in Python.  This probably stems from the fact I
tend to program in a sort of 'side-effectless style' which Python
encourages when dealing with the imperitive programming style (as
differing from the object-oriented style one gets into which hinges on
self-encapsulated objects and instances).  That a language encourages
one style of programming at the expense of another is hardly reason to
change the language, /especially/ if both idioms are perfectly valid
means within the language.

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From kernr at mail.ncifcrf.gov  Thu Aug 19 12:45:08 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Thu, 19 Aug 1999 16:45:08 GMT
Subject: NEWBIE: Setting WinNT python to point to my tcl DLL
References: <37BC1801.3404C9E8@nortelnetworks.com>
Message-ID: <37bc33e0.3846497@news.erols.com>

On Thu, 19 Aug 1999 09:43:13 -0500, "Ron A. Zajac"
<zajac at nortelnetworks.com> wrote:

>
>Dear Python Anciens,
>
>  I've installed the latest Python on my WinNT system, but declined the
>Tk install, since I already have tcl/Tk8.1.1 on my system.  It appears
>to be the case I need to point the system to my current tcl/Tk install,
>however.  "IDLE" can't find "tcl80.dll" which, of course, doesn't exist.
>
>  Where can I make this change so IDLE can find the relevant DLL?

You can't.  IDLE itself isn't searching for tcl80.dll; the _tkinter
extension module is.  Someone *might* have Windows binaries of
_tkinter.pyd that link with TCL/TK 8.1.1, but you probably won't have
much luck.

My suggestion is to install TCL/TK 8.0.x from the Python installer.  I
don't know what uninstallation you would have to do beforehand,
though.  Do you really need TCL/TK 8.1.1 instead of 8.0.x?

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From bowman at montana.com  Sun Aug 22 13:52:13 1999
From: bowman at montana.com (bowman)
Date: Sun, 22 Aug 1999 11:52:13 -0600
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <37bc313b@cs.colorado.edu> <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us> <019301beeaea$004c4d80$f29b12c2@secret.pythonware.com> <37bd4876@cs.colorado.edu> <MPG.1229cd5eed750a81989687@news.mop.no>
Message-ID: <37C038CD.5BE69AC7@montana.com>

Alexander Staubo wrote:
> 
> Your tool isn't everybody else's tool. Personally I don't even know what
> "apropos" is. Even if I made an effort at using it, it doubt it would
> make _my_ workday any easier;

'apropos' is the tool that saves man's bacon. You're sitting there
trying to find some info on a widget, but you can't remember if it is
'button', pushbutton', 'Button', 'XmpushButton', 'XmPushButton',
'TkButton'  blah, blah, blah, so you type

apropos button

It also solves the problem of 'foo' having entries in man 1, man 3, man
n.

If you have to use the antiquated, deprecated,
no-longer-supported-see-Info, man pages, it is an invaluable tool.

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From chad at vision.arc.nasa.gov  Wed Aug 11 22:11:56 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Thu, 12 Aug 1999 02:11:56 +0000
Subject: Problem when using FileDialog from PMW Dialog.  Quick fixes?
References: <37B21CE1.488F45FA@vision.arc.nasa.gov>
Message-ID: <37B22D6C.4ADD0B18@vision.arc.nasa.gov>

Chad Netzer wrote:

> Before I pursue this further, can anyone tell me if it is even worth
> trying?  It is nice to use the canned FileDialog, but its implementation
> is somewhat crufty (IMHO) and I can probably work around it in other
> ways (Creating my own FileDialog from PMW widgets is a nice possibility.

Yet another reply to my own post.  After this last message, I said "What
the hell",
and converted the FileDialog.py from the standard Python Tkinter
distribution
to use PMW dialogs.  It took about half an hour :)  Damn I love this
language!

I'm on a bit of a deadline so I hadn't thought to devote time to this
originally, but
it was quick, easy, the result looks nicer, and it WORKS.  I'll post it
somewhere
sometime later, and hopefully reorganize it to fit the general PMW scheme
of
organizing objects, then submit it for inclusion.

Anyway, I'd still like to hear an answer to my original query, but for now,
it is
academic (for me).

Cheers,
Chad
chad at vision.arc.nasa.gov





From chad at vision.arc.nasa.gov  Thu Aug 19 22:07:58 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Fri, 20 Aug 1999 02:07:58 +0000
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <8E27B3FC4duncanrcpcouk@news.rmplc.co.uk> <37bc4a7d@cs.colorado.edu>
Message-ID: <37BCB87E.A95FC90@vision.arc.nasa.gov>

Tom Christiansen wrote:

> You're right.  I should have written
>
>     new = []
>     for i in range(0,len(old)):
>         new[i] = old[i]
>
> And that certainly is parallel to the dictionary case.  Dictionary
> auto-allocate.  Lists don't.

Again, why not just use new.append(old[i])?  If you REALLY want to
write it with an equals sign or start the indexing at other than zero, then
declare new as a dictionary (integers are valid keys since they are non
mutable)

old = [1,2,3,4,5,6,7,8,9]
new={}
for i in range(4, len(old)):
   new[i] = old[i]

Autoallocating arrays, or arrays with holes, are simply Python dicts.
Of course, having something like new.sortedprint() might be nice,
but it is easy to define a function.

def sortedprint(dict):
   spam = dict.keys()
   spam.sort()    # What, you've never heard of the amazing spamsort?
   for i in spam:
      print dict[i],
   print


Chad Netzer
chad at vision.arc.nasa.gov





From skip at mojam.com  Fri Aug 20 12:40:15 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 20 Aug 1999 11:40:15 -0500 (CDT)
Subject: making python man pages looks hard
In-Reply-To: <37bd0b97@cs.colorado.edu>
References: <1103_935134343@shoal>
	<37bd0b97@cs.colorado.edu>
Message-ID: <14269.33828.537574.712197@dolphin.mojam.com>

    Tom> Try

    Tom>     http://search.cpan.org/

Tom,

Just out of curiosity, how is CPAN populated and kept in sync?  I went the
the Winnipeg site at

    http://theory.uwinnipeg.ca/search/cpan-search.html

and searched for Frontier (Ken MacLeod's Perl XML-RPC package) and found a
couple links to a readme and a tgz.  Clickly the readme link failed the
first time, but succeeded the second (guess it was choosing different
mirrors and got a hoser the first time).

I then tried the same search at search.cpan.org and got no relevant hits.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From moshez at server.python.net  Tue Aug 31 01:50:42 1999
From: moshez at server.python.net (Moshe Zadka)
Date: Tue, 31 Aug 1999 01:50:42 -0400 (EDT)
Subject: 'if foo' calls getattr?!?
In-Reply-To: <Pine.LNX.4.10.9908310124220.9802-100000@starship.python.net>
Message-ID: <Pine.LNX.4.10.9908310132450.9802-100000@starship.python.net>

On Tue, 31 Aug 1999, Moshe Zadka wrote:

> Bzzt! foo!=None is slower then foo is not None, but not by much: it will
> first check whether type(foo) is not type(None), so it is pretty fast.
> 
> truth-and-nothing-are-always-popular-subjects-ly y'rs, Z.

Sorry to reply to my own posts, but there was a mistake: foor!=None will
make a call to __getattr__ for __cmp__, so it will be very slow.

oops-ly y'rs, Z.




From paul at prescod.net  Sat Aug 21 08:42:19 1999
From: paul at prescod.net (Paul Prescod)
Date: Sat, 21 Aug 1999 12:42:19 GMT
Subject: Man pages would be cool.
References: <37BC6986.A66B044E@starmedia.net>
				<199908191634.KAA01333@jhereg.perl.com>
				<14268.21198.934117.241668@anthem.cnri.reston.va.us>
				<14268.23007.239507.166209@weyr.cnri.reston.va.us>
		<y0j7lmq44wd.fsf@vier.idi.ntnu.no>
		<37BDD1B9.99588CB@prescod.net> <37BE24E0.79EA064E@montana.com>
Message-ID: <37BE9EAB.CB7B1380@prescod.net>

bowman wrote:
> 
> I'll pass on the man pages noise, but it is easy enough to add IDLE to
> the menus. In fact, it fits very nicely in the slot where xeyes used to
> live :)

I'm talking more from a marketing perspective. Wouldn't it be great if
new Linux users just walking through the menus would "stumble" upon
IDLE? I think that it would be right in Linux's mandate to "educate the
masses" and Python's mandate to make programming accessible to the
masses.

 Paul Prescod



From tim_one at email.msn.com  Tue Aug 17 21:20:16 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 21:20:16 -0400
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <wkyafagqey.fsf@turangalila.harmonixmusic.com>
Message-ID: <000501bee917$d55fd660$eaa22299@tim>

> | def randomize(a, b):
> |     return random.choice([-1, 0, 1])
> |
> | l = range(2, 20)
> | l.sort(randomize)

[Dan Schmidt]
> I'm not sure how robust this is.  I've seen at least one implementation
> of qsort crash when given an inconsistent comparison function (e.g.,
> one that doesn't obey transitivity).

Python implements its own sort function (in 1.5.2 it's not quicksort, btw,
but a hybrid of common special cases, binary insertion and samplesort).  A
very long time ago it used the platform qsort, but that blew up on some
systems; iirc, usually due to qsort being implemented non-reentrantly.  The
current sort is robust against all that kind of stuff, including insane
comparison functions; it always returns *some* permutation of its input.

> And how is sort() supposed to know when it's done?

Surprisingly simple <wink>:

			/* Find another slice to work on. */
			if (--top < 0)
				break;   /* no more -- done! */

insane-comparators-simply-yield-insane-meanings-for-"sorted"-ly
    y'rs  - tim





From howard.s.modell at boeing.com  Thu Aug 19 18:16:01 1999
From: howard.s.modell at boeing.com (Howard S.Modell)
Date: Thu, 19 Aug 1999 22:16:01 GMT
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk> <19990816205639474.AAA315.413@max41127.izone.net.au> <lDvYnDAz1Hu3EwB6@jessikat.demon.co.uk> <37BB2E1D.C0733F85@boeing.com> <7pgd1g$hj6$1@newshost.accu.uu.nl>
Message-ID: <37BC8221.654373C8@boeing.com>

Martijn Faassen wrote:
> 
> Howard S.Modell <howard.s.modell at boeing.com> wrote:
> [snip[
> > Perl? Python?  If we all programmed in SNOBOL4 the way g-d intended,
> > we wouldn't have to worry about debates like that!
> 
> Funny, I read that 'the way g-d indented'. :)
> 
> Regards,
> 
> Martijn

I would assume that when g-d writes python code, it just naturally
falls into divine alignment, no spaces or tabs necessary ...

-- 
Perl? Python?  If we all programmed in SNOBOL4 the way g-d intended, 
we wouldn't have to worry about debates like that!
---------------------------------------------------------------------
Howard S. Modell ???                    howard.s.modell-at-boeing.com



From kas at maps.magnetic-ink.dk  Sun Aug  8 01:10:52 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Sun, 8 Aug 1999 07:10:52 +0200
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk>
Message-ID: <slrn7qq4as.cc2.kas@titan.magnetic-ink.dk>

Phil Hunt wrote:

>    #bak.py
>    from hhh import *
>
> [...]
> 
>    #hhh.py
>    from bak import *

Do we have a circular reference here?


  //Klaus

-- 
???[ Magnetic Ink ]????????????????????????????????????????????????? ><> ???
???[ http://www.magnetic-ink.dk/ ]??????????????????????????????????????????



From skaller at maxtal.com.au  Wed Aug  4 02:36:50 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Wed, 04 Aug 1999 06:36:50 GMT
Subject: 1.5.2 for: else:
References: <1279007788-11957995@hypernet.com> <slrn7psisn.da.wtanksle@dolphin.openprojects.net> <000c01bed893$2ce8f7d0$f29b12c2@secret.pythonware.com> <slrn7pum9t.2ro.wtanksle@dolphin.openprojects.net> <7nnuda$u47@news.dns.microsoft.com>
Message-ID: <37a7de3f.1804238928@news.triode.net.au>

On Wed, 28 Jul 1999 14:57:55 -0700, "Vadim Chugunov" <chega_ at yahoo.com> wrote:

>When I first saw Python syntax for loops I said: "A-ha!  So in Python I will not
>need a goto
>in situations like this:
>-----
>for(Item* pitem=pseq->First(); pitem; pitem=pitem->Next())
>    if (pitem->key==42)
>        goto Found;
>
>pitem = pseq->Insert(new Item());
>Found:
>    // use item pointed to by pitem
>-----
>In fact, I do not see any other good use for else: clause in a loop.

Isn't this a fine reason for it?
John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From jstevens at bamboo.verinet.com  Sat Aug 14 10:57:21 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: Sat, 14 Aug 1999 14:57:21 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ce93@cs.colorado.edu> <slrn7r9la9.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nde.h3c.jstevens@bamboo.verinet.com> <slrn7r9p2h.pu8.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <slrn7rb0uh.cf9.jstevens@bamboo.verinet.com>

On 14 Aug 1999 03:36:23 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>On Sat, 14 Aug 1999 03:08:33 GMT,
>    John Stevens <jstevens at bamboo.verinet.com> wrote:
>>On 14 Aug 1999 02:32:12 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:
>>>On 13 Aug 1999 20:04:03 -0700,
>    John W. Stevens <jstevens at basho.fc.hp.com> wrote:
>>>>
>>>>Perl doesn't have lists.  Python doesn't have built-in arrays.
>>>
>>>You should learn some perl you now..
>>>
>>>@array = (1,10,20,30);
>>>$from_list = (1,10,20,30);
>>>$from_array = @array;
>>>print "$from_list\n$from_array\n";
>>>
>>>Will output :
>>>30
>>>4
>>
>>The @ prefix denotes an array.  You, yourself, should learn
>>Perl.  Calling an array a list, doesn't make it one.
>
>Can you read?
>
>Can you see a @ in the following line of code :
>
>$from_list = (1,10,20,30);

Your example included:

@array = (1,10,20,30);
$from_list = (1,10,20,30);
$from_array = @array;
print "$from_list\n$from_array\n";

The line you specify does not contain a list, it contains a
tuple.

>No you can't.  Is there an array in that line of code?

No, and that isn't a list, even if you call it one.  It is a tuple.

>No. Is there a list
>in that line of code?

No, it's a tuple, not a list.

>Yes.

No.  Calling it a list, doesn't make it a list.

To be able to perform operations on the above, you'd have to
assign the values of the tuple to an array.

In which case you'd be performing array operations, not list
operations.

>>Not built in, it doesn't, unless you define array and list as being
>>different words for exactly the same type/class.
>
>I don't think you can get more built in then perl lists.

You get tuples in Perl.  The book calls then lists, but you cannot
perform any of the standard, defined list operations on a Perl
"list".

All "list" operations are performed by shoving "lists" into
array's.

Go ahead, show me a variable that contains a list.  In your
example:

$from_list = (1,10,20,30);

$from_list isn't a list.

>Again I repeat, here is a perl list : ('a', 'b', 'c') or qw(a b c)

Again, you are wrong.  The construct ('a', 'b', 'c') is a tuple.
Perl'ers may call it a list, but how do you perform a an insert
on the above?

You don't.  You copy the contents of your tuple into an array,
then perform array operations on that array.

>>I know Perl.  You need to learn Python.
>
>Did I mention a single bit of python syntax in my post? No. 

So what?  I can still make a comment.  And if you had been
paying attention, you would have realized the context of this
thread.

>What was the point of saying I need to learn Python?

Because, in the words of TC, "Learning is a good thing".

More specifically, had you studied Python, it would have been
clear that what Perl calls a list is more properly refered to
as a tuple.

>You might as well have told me to learn C or Lisp.

Yes.  You should.  Had you learned Lisp, list processing would
be second nature to you, and you would have a clearer idea between
"lists" and "tuples" and "arrays".

>>If @ denotes list, then the following Perl would be illegal:
>>
>>@ary = (1, 2, 3);
>>@ary[5] = "Test";
>>
>>But, obviously, this is not illegal.
>
>@ary[5] is an array slice.

Yes.  So what?

>It is an array. It happens to be an array 
>that constists of only one element.

Again, so what?

>Why would it be illegal?

If @ary were a list, it would be illegal.

>It is just
>the special case of :

Stop.  You are talking about arrays, not lists.  I repeat, if
@ary were a list, the above operation would be illegal.

>@ary[1,2,3,4,5,10,20]
>
>It is an array. Thus it has a @.

Yes.  It is an array, not a list.  Though to be perfectly anal,
it might be best to refer to that as a vector. . . ;->

>>Really?  What is the prefix character that denotes a list?
>
>You can't have a list in a variable.

Correct.  So, Perl does not have lists, it has tuples.

>Again here is a list :
>
>(1,2,3,4,5,6,7,8)

No, that is a tuple.

>push( (1,2,3,4,5), 6); 
>
>This results in an error message, funnily enough the message is :
>Type of arg 1 to push must be an array (not list)

Yes.  But calling a tuple a list (or for that matter, calling an
array a list) does not make it one.

Had that been a real list, and assuming that "push" really means,
append, that should have been legal.

>Still claim perl has no lists?

Yes.

>Of course you know better then perl itself.

In this case, I would say that Perl should use the same terminology
as other fields of study.  So, yes, I would consider that Perl
makes a mistake, calling this a list.

John S.



From meowing at banet.net  Thu Aug  5 22:49:22 1999
From: meowing at banet.net (meow)
Date: 05 Aug 1999 22:49:22 -0400
Subject: Multi-part question: /etc/passwd and /etc/shadow   and   how to load   modules with a full path
References: <7odft3$got$0@199.201.191.2>
Message-ID: <87672ttyh9.fsf@slip129-37-108-208.nj.us.ibm.net>

Coltrey Mather <Sinistrad at penguinpowered.com> wrote:

>     Are there any modules that could be easily incorporated into a
> program to get the password for a user from /etc/shadow (e.g.: for user
> authentication in a server process) that would check to see if the
> password given for the username given is the right combination?

Yeah, the pwd module is generally going to get its info from
/etc/passwd, which isn't usually what you want these days.

I don't know if someone's written a module to get shadow info, but let
me try to talk you out of doing this directly.  Reading the shadow
file generally requires being root or in a privileged group, so you
want to isolate that from user apps as much as you can.  You'll also
run into portability problems -- not all shadow systems use the same
API, not all password files use crypt(), and some systems out there
are still running without shadow passwords.  In a word, yecch.

One sorta neat and easy safe-ish method I've seen to get this info
(while leaving the safety problems to others) is to run a POP3 server
on the host in question, and use that to handle the password ugliness.
Just see if you can log in, skip doing silly things like checking for
mail.  The standard poplib module will work just fine here.  And, it
will even work on systems that don't run UNIX.

Or, you could write a tiny separate setuid program (daemon or not,
your call) to do just the password checking part.

> Also, if I want load a module, /home/me/mymodule.py, from a python
> program, /usr/local/bin/thisprogram.py, how do I go about doing that?

One way would be:

import sys
sys.path.insert(0, '/home/me')
import mymodule



From s323140 at student.uq.edu.au  Wed Aug 18 09:45:32 1999
From: s323140 at student.uq.edu.au (Rob Hodges)
Date: 18 Aug 1999 23:45:32 +1000
Subject: waiting on a callback with PyGTK
Message-ID: <m37lmtb3sz.fsf@phaedrus.zen.xom>

Hello Python gurus,

Please forgive me if this is a really dumb question!

I've been playing around with PyGTK and have a little program that
pops up a dialog with a text entry in it.  When the user hits Enter in
the text entry, a callback grabs the contents of the entry.

Now here's my problem: I want the program to sit there doing
essentially nothing until the callback returns.  I thought I'd be able
to do this by inspecting the variable that the callback sets to the
contents of the entry: Initialise the variable to an empty string or
None, pop up the dialog and then inspect the variable (member 'text' of
class instance 'dialog') like this:

while not dialog.text:
        time.sleep(0.02)

But then the dialog just never comes up, even though the inspection
comes after the creation and show()ing of the dialog.  What is
happening here?  And more to the point, how do I get around it?  I'm
sure this must be a very common thing to do and I'm missing something
basic.

-Rob




From bowman at montana.com  Sat Aug 28 10:11:11 1999
From: bowman at montana.com (bowman)
Date: Sat, 28 Aug 1999 08:11:11 -0600
Subject: A Plea For A Return To Sanity!!!
References: <000301bef12c$1bfd6940$ad2d153f@tim>
Message-ID: <37C7EDFF.80FEB1E9@montana.com>

Tim Peters wrote:
> 
> > Most perlers I have met are a preaty logical bunch,
> 
> Then you'll find Python a great relief!  Logic doesn't impress anyone here;
> paying your PSA dues is all that counts.

I don't have the chattier first edition of Programming Perl at hand, but
didn't Larry Wall mention logic and consistency make life boring, or
something like that?


-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From bitwise at io.com  Tue Aug 24 14:35:14 1999
From: bitwise at io.com (Bit Nik)
Date: Tue, 24 Aug 1999 13:35:14 -0500 (CDT)
Subject: Python equivilent of pack()?
Message-ID: <Pine.LNX.4.10.9908241328450.5053-100000@eris.io.com>

I'm writing a small program that needs to be able to decode a RADIUS
packet obtained from a snoop trace. I'd like to do it in Python 'cause its
just such a darn handy little programming language.

My question is this: is there a functional equivilent to the Perl pack()
and unpack() functions in Python? Or even better, is there an easier way
to do what I want with Python? If not, how many people think this would be
a useful addition to the Python standard library?

Thanks in advance.

ACT




From skip at mojam.com  Tue Aug 17 10:57:01 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 17 Aug 1999 09:57:01 -0500 (CDT)
Subject: if you want POSIX, import posix, not os
In-Reply-To: <00c301bee884$42b1d3b0$f29b12c2@secret.pythonware.com>
References: <37B8EC03.835F8A51@w3.org>
	<00c301bee884$42b1d3b0$f29b12c2@secret.pythonware.com>
Message-ID: <14265.30090.868708.792121@dolphin.mojam.com>

    Fredrik> in fact, the "os" module docs says:

    Fredrik>     This module provides a more portable way of using
    Fredrik>     operating system (OS) dependent functionality than
    Fredrik>     importing an OS dependent built-in module like posix
    Fredrik>     or nt.

I agree with Dan that "posix" should imply "portable".  Outside the Python
world the name "POSIX" has some (not many perhaps, but some) cross-platform
portability connotations.  Perhaps the posix module should be renamed to
"unix".  That way, if you wanted os-specific functionality, you'd import
"unix" or "mac" or "nt" or "dos".

If we had a unix module, each os-dependent module could be split into two
chunks, "posix-whatever" and "non-posix-whatever", say "posixunix" and
"nonposixunix".  On my Linux box, I could import posix to get the portable
subset or import unix to get posix and all the unix-specific stuff.  The
posix module would know to import "posixunix" or "posixnt" or ... as
appropriate.  The unix module would be nothing more than a stub containing

    from posixunix import *
    from nonposixunix import *

All the various "posixwhatever" modules would be required to implement the
same functionality.  If I wanted to add a new function to the unix module it
would have to go into nonposixunix unless I could implement it in all the
other posix* modules.

<nomex state=on>

perhaps-we-can-let-posixdos-be-more-feeble-ly yr,s

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From fdrake at acm.org  Thu Aug 19 14:25:27 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 19 Aug 1999 18:25:27 GMT
Subject: Python to Perl Conversions
In-Reply-To: <slrn7rog4u.hri.jblaine@shell2.shore.net>
References: <37bc1303@cs.colorado.edu>
	<slrn7ro9ns.h6v.jblaine@shell2.shore.net>
	<37bc2ee5@cs.colorado.edu>
	<slrn7rog4u.hri.jblaine@shell2.shore.net>
Message-ID: <14268.19479.608776.126088@weyr.cnri.reston.va.us>

Jeff Blaine writes:
 > can certainly agree that SOME sort of man page for 'python.1' would
 > make some sense, and more than that would only make more people happy...

  'man python' works for me; it's installed by default.  You can use
all the usual manpage handling tools with it, too, including troff &
apropos.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From mjboyles at voyager.vrve.iupui.edu  Mon Aug 16 12:25:51 1999
From: mjboyles at voyager.vrve.iupui.edu (Michael Boyles)
Date: Mon, 16 Aug 1999 11:25:51 -0500 (EST)
Subject: Problem with "open"...
Message-ID: <Pine.SGI.3.96.990816111849.37560A-100000@voyager.vrve.iupui.edu>

I am a very new user to python, but I thought I picked it up pretty
quickly. The code I have been using has been working for several
months now. I recently reorganized the modules (placement in the .py
file) and renamed some variables (nothing fancy) and I now get an
error from the following code:


class effect:
  ...
	
  def render():
    ...
    self.createViewFile("0000.vf")
    ...

  def createViewFile(self, filename):
    f = open(filename, "w")
    f.write("rview")
    f.close()
	
  ...

	
The error is on the "open" function. It reads:
	
  Exception in Tkinter callback
  Traceback (innermost last):
  File "/usr/local/lib/python1.5/lib-tk/Tkinter.py", line 726, in __call__
    return apply(self.func, args)
  File "/usr/people/mjboyles/effect/bin/effectGUI.py", line 156, in render
    self.menuListener.render()
  File "effect.py", line 423, in render
    f = open(("%s.vf" % name), 'w')
  TypeError: illegal argument type for built-in operation


Any suggestions? I have no idea about the error. When I try the
function during regular interperator mode from the prompt, it works
fine. Is there a description about the possible python errors
somewhere? 

Thanks in advance...
- Mike
   ___________________________________________________
 _/                                                   \_
|  Michael Boyles                  Research Programmer  |
|  Advanced Visualization Lab       Indiana University  |
|_ mjboyles at iupui.edu     www.vrve.iupui.edu/~mjboyles _|
  \___________________________________________________/




From tim_one at email.msn.com  Sat Aug 28 04:05:21 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 28 Aug 1999 04:05:21 -0400
Subject: Fuzzy string matching?
In-Reply-To: <14279.9340.845977.120321@dolphin.mojam.com>
Message-ID: <000001bef12c$139c0680$ad2d153f@tim>

[John Machin, disses Python's soundex implementation]

[Skip Montanaro]
> ...
> Guido would like sondex.c out of the core.  Tim Peters sent a Python
> replacement to me.  I was supposed to give it some life.  It's still
> sitting in my inbox.

So what does it need to come to life?  It's code; it works <wink>.

> I'll try to take care of it next week (bug fixes and all).

It didn't have the bugs John's worried about.

> (I realize that doesn't make Soundex any better for the fuzzy
> matching algorithm, but I should at least fix any obvious bugs...)

I find Knuth's description of the algorithm ambiguous in several respects, and
there are many incompatible soundex algorithms "out there".  The one I sent
before probably should have treated non-letters as if they didn't exist, so
I'll attach a changed version that does so.

soundex-is-a-pragmatic-hack-so-right-vs-wrong-is-fuzzy-ly y'rs  - tim

NDIGITS = 3

import string
_upper = string.upper
_translate = string.translate

# P for Punctuation -- all characters treated as if they weren't there
_tran = ["P"] * 256
def _setcode(letters, value):
    for ch in letters:
        _tran[ord(_upper(ch))] = _tran[ord(ch)] = value
_setcode("bfpv", "1")
_setcode("cgjkqsxz", "2")
_setcode("dt", "3")
_setcode("l", "4")
_setcode("mn", "5")
_setcode("r", "6")
# B for Break -- these guys break runs
_setcode("aeiouy", "B")
# I for Invisible -- they're invisible except as first char
_setcode("hw", "I")

_punctuation = filter(lambda i: _tran[i] == "P", range(256))
_punctuation = string.join(map(chr, _punctuation), "")
assert len(_punctuation) == 256 - 52, "Soundex initialization failed"

_tran = string.join(_tran, "")
del string, _setcode

def soundex(name):
    """name -> Soundex code, following Knuth Vol 3 Ed 2 pg 394."""

    while name and _tran[ord(name[0])] == "P":
        name = name[1:]
    if not name:
        raise ValueError("soundex requires non-empty name argument")
    coded = _translate(name, _tran, _punctuation)
    out = _upper(name[0])
    lastrealcode = coded[0]
    ignore_same = 1
    for thiscode in coded[1:]:
        if thiscode == "B":
            ignore_same = 0
            continue
        if thiscode == "I":
            continue
        if ignore_same and lastrealcode == thiscode:
            continue
        out = out + thiscode
        lastrealcode = thiscode
        ignore_same = 1
        if len(out) > NDIGITS:
            break
    if len(out) < NDIGITS + 1:
        out = out + "0" * (NDIGITS + 1 - len(out))
    return out

def _test():
    global nerrors
    def check(name, expected):
        global nerrors
        got = soundex(name)
        if got != expected:
            nerrors = nerrors + 1
            print "error in Soundex test: name", name, \
                  "expected", expected, "got", got
    nerrors = 0
    check("Euler", "E460")
    check("Ellery", "E460")
    check("guass", "G200")
    check("gauss", "G200")
    check("Ghosh", "G200")
    check("HILBERT", "H416")
    check("Heilbronn", "H416")
    check("Knuth", "K530")
    check("K ** n  U123t9247h   ", "K530")
    check("!Kant", "K530")
    check("Lloyd", "L300")
    check("Liddy", "L300")
    check("Lukasiewicz", "L222")
    check("Lissajous", "L222")
    check("Wachs", "W200")
    check("Waugh", "W200")
    check("HYHYH", "H000")
    check("kkkkkkkwwwwkkkkkhhhhhkkkkemmnmnhmn", "K500")
    check("Rogers", "R262")
    check("Rodgers", "R326")
    check("Sinclair", "S524")
    check("St. Clair", "S324")
    check("Tchebysheff", "T212")
    check("@@=T-c=h,e+b#y{s}h\000e\377f:f|", "T212")
    check(" C h e b y s h e v ", "C121")

    if nerrors:
        raise SystemError("soundex test failed with " + `nerrors` +
                          " errors")

if __name__ == "__main__":
    _test()



a


From thantos at brimstone.mecha  Sun Aug  8 22:54:18 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Mon, 09 Aug 1999 02:54:18 GMT
Subject: lambda fun: infinite streams
References: <37ABADB1.834A664E@go2net.com> <slrn7qnsod.jjr.thantos@brimstone.mecha>  <y0jlnbmgzis.fsf@drue.idi.ntnu.no>
Message-ID: <slrn7qsger.d0.thantos@brimstone.mecha>

On 08 Aug 1999 15:42:03 +0200, Magnus L. Hetland <mlh at idt.ntnu.no> wrote:
>Well... You could make one yourself...

True, but it wouldn't have the speed of the built-in map implimented
in C, would it?  :)

(Yes, longing for such things is left over from tinkering with
Haskell; there are so many algorithms that fall to simple expression
applied to Streams and lazy-evaluation.)

>def mapStream(f,s):
>    return (apply(f,[head(s)]),
>           lambda f=f, s=s: mapStream(f,tail(s)))

Hmmmmm, that should certainly work ...

I'm tempted to try and get Stackless Python working with this just to
see what kind of interaction it'd have.

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From chinaski_py at my-deja.com  Tue Aug 24 04:34:06 1999
From: chinaski_py at my-deja.com (chinaski_py at my-deja.com)
Date: Tue, 24 Aug 1999 08:34:06 GMT
Subject: NEWBIE:Let's try that again..Help w/path
References: <7pm0o4$e0h$1@nnrp1.deja.com> <7pshkp$q47$1@nnrp1.deja.com>
Message-ID: <7ptldu$jfv$1@nnrp1.deja.com>

In article <7pshkp$q47$1 at nnrp1.deja.com>,
  chinaski_py at my-deja.com wrote:
> In article <7pm0o4$e0h$1 at nnrp1.deja.com>,
>   chinaski_py at my-deja.com wrote:
> > I'm new and need help with my Python path.When I go to open a Python
> > file (.py) it won't open .I go to (on Win98):
> > start
> > run
> > sysedit
> > and here is my path:
> > path %path%;C:\Perl\bin;path=C:\PROGRA~1\Python\;%path%
> >                 Any help on this would be appreciated,
> >                                 G.Norton (chinaski_py)
> >
> > Sent via Deja.com http://www.deja.com/


> > Share what you know. Learn what you don't.
> >
> Thomas,
>       tried your path but can't open "IDLE".The icon is there in the
> file, but I click and nothing happens.This is what I get:
>                        PROGRAM NOT FOUND
>                  This program is needed for opening files of type
>                      'Python File' (no console)
> shouldn't it be in the path? Glenn (chinaski)
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>

  PROBLEMS SOLVED!!!!!!!! THANKS!!!!!!!!!!!!!!!!!!!!!!!!! 8/24/99


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From stalnaker at acm.org  Thu Aug 12 13:20:24 1999
From: stalnaker at acm.org (Max M. Stalnaker)
Date: Thu, 12 Aug 1999 10:20:24 -0700
Subject: win2k python install
Message-ID: <37B30258.6D3B816C@acm.org>

Installing python on a win2k RC1 box, I had the expectation that the
python executable would be on the path.  Thus, from an arbitrary
directory, with a dos box,  python test.py would find python and execute
it.  This does not happen and I am wondering if the install is
*supposed* to put the python executable on the path.




From jim at interet.com  Thu Aug 26 08:48:31 1999
From: jim at interet.com (James C. Ahlstrom)
Date: Thu, 26 Aug 1999 08:48:31 -0400
Subject: Making A Standalone Executable Under Windows NT/95/98
References: <hUQw3.117$Kl.24641667@speed.city-net.com>
Message-ID: <37C5379F.86D73C8E@interet.com>

"jhomme.." wrote:
> 
> Is this possible?  If so, How?

One way is to use "freeze" in ./Tools/freeze to convert
all your *.pyc and needed library files into C strings.
Then re-link Python with this module.  If your main
program is included with the name "__main__", it will
be started.

Also check out Gordon McMillan's win32 installer
    http://starship.python.net/crew/gmcm/install.html.

Jim Ahlstrom



From glen at electricorb.com  Sat Aug 28 15:07:32 1999
From: glen at electricorb.com (Glen Starchman)
Date: Sat, 28 Aug 1999 12:07:32 -0700
Subject: VB to Python Converter
Message-ID: <37C83374.B2CC17B8@electricorb.com>

I am in the preliminary stages of creating a VB(actually ASP written in
VBScript)  to Python converter and am about to create the VBScript
grammar. Before I go down this painful road, I am wondering if anyone
happens to have any leads on such a beast. I am not exactly grammar
savvy and there are quite a few little interesting pitfalls in
VB(script).

I have worked out some theories on how to handle built-in (or at least
Server.CreateObject(x) ) instances... for example I have a module that
mimics the ADO object model fairly well... ditto for Request, Response,
etc...

Any help is MUCH appreciated!






From skaller at maxtal.com.au  Sun Aug 15 11:28:15 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sun, 15 Aug 1999 15:28:15 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <934326918snz@vision25.demon.co.uk>
Message-ID: <37b6d971.341603770@news.triode.net.au>

On Tue, 10 Aug 99 23:15:18 GMT, philh at vision25.demon.co.uk (Phil Hunt) wrote:

>In article <37b075ae.2367293988 at news.triode.net.au>
>           skaller at maxtal.com.au "John (Max" writes:
>> This is a request for comments on Viper,
>> Extensions:
>>         6) what do you want?
>
>C-style comments

	With what syntax, and why?

1. Syntax:

The C syntax doesn't support nesting; OTOH the PL/1 ML style 
(* ... *) syntax does.

	New lexicology for strings and comments is easy
enough to add, since it only affects the lexer. However, it is
also easy to create an unreadable mess if _too many_
new lexical forms are introduced.

2. Why?

One obvious reason: annotating things like function parameters
with C++/Python style comments can expand source code too much:

	def f( # a function
		a, # arg1
		b # arg2
	):

might be written

	def f /* a function */ ( a /* arg1 */, b /* arg2 */ ):

on one line with C style comments. 

>sorted dictionary loops, eg:
>for k, v in dict.sortedValues():
>where this is sorted by keys.

	Something to support dictionary iteration is
needed. My thought was:

	for k in dict: value = dict[k] ...

which is equivalent to

	keys = dict.keys()
	keys.sort()
	for k in keys: value = dict[k]

Unfortunately, you can't write

	for k in dict.keys().sort():

because the keys() refers directly to the dictionary
keys, and sort() sorts in place. Both these methods
are the way they are for efficiency. Perhaps a new
method would be a good solution (for Python 1.6 too):

	for k in dict.sorted_keys(): ...

however, there is a related idea here, that a dictionary
with all values None is isomorphic to a set. Thus

	k in dict

would mean

	k in set

which is natural notation, and 

	k in dict 

is an extension to dictionaries in which the
set is the set of keys.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From paul at prescod.net  Fri Aug 27 13:43:06 1999
From: paul at prescod.net (Paul Prescod)
Date: Fri, 27 Aug 1999 17:43:06 GMT
Subject: Python Tools Make a Strong Showing
Message-ID: <37C6CE2A.BDAB1FD0@prescod.net>

http://www.xml.com/pub/1999/08/excelon/montreal.html#python

 Paul Prescod



From do826 at micsv.cwb.gov.tw  Mon Aug 23 23:24:04 1999
From: do826 at micsv.cwb.gov.tw (Alex)
Date: Tue, 24 Aug 1999 11:24:04 +0800
Subject: Recommand a good book for learning Python
Message-ID: <7pt2oo$rts@netnews.hinet.net>

Hi all:

I want to learn Python Programming language, but I don't know which book is
better for a newbie. So I need some suggestions. Thanks a lot.

P.S: Whould you like to reply one to my E-Mail?

Regards

--Alex





From Svante.Kleist at NEMESIS.se  Mon Aug 30 09:14:19 1999
From: Svante.Kleist at NEMESIS.se (Svante Kleist)
Date: Mon, 30 Aug 1999 15:14:19 +0200
Subject: IDLE: What to use instead of sys.argv[0] ?
Message-ID: <MPG.1231b2f7edc3c9eb989681@news.pi.se>

Dear Pythoneers,

Normally, I can retrieve the path and file of the executing Python script
from sys.argv[0]. But when running a script inside IDLE, this gives e.g.

	E:\local\Python\Tools\idle\idle.pyw

Is there a way to get the path/file to the script I'm executing in IDLE?

Thanks. 

/ Svante Kleist - NEMESIS systemDesign, Stockholm



From kernr at mail.ncifcrf.gov  Tue Aug 17 20:45:16 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Wed, 18 Aug 1999 00:45:16 GMT
Subject: Wrappers to FreeType
Message-ID: <37b9ff95.10762600@news.erols.com>

I am in the middle (well, nearer to the end) of SWIGging the FreeType
library <URL: http://www.freetype.org>.  FreeType is an Open Source
(with an almost Python-style licence) library for rendering Truetype
fonts onto bitmaps (or pixmaps).

My questions are:

  * Is anyone else interested in this?

  * Would they (if they, indeed, do exist) mind requiring Numeric to 
    provide the arrays that the engine rasters into?  SWIG provides
    the ability to make the necessary arrays, but accessing them from
    Python is slow.  Rastering into a Numeric array would instantly 
    provide the data to Python (and Python-level interaction with PIL
    is fast, too).

  * Since interaction with PIL at the C level would be great and I
    don't know squat about how to make that happen, would anyone care
    to contribute code (or pointers) for that?

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From mal at lemburg.com  Fri Aug  6 05:16:46 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 06 Aug 1999 11:16:46 +0200
Subject: [Image-SIG] Proper application of the buffer interface
References: <Pine.WNT.4.05.9908042019270.40-100000@david.ski.org> <14249.43677.175790.689266@weyr.cnri.reston.va.us>
Message-ID: <37AAA7FE.FFE0FD2@lemburg.com>

Fred L. Drake, Jr. wrote:
> 
> David Ascher writes:
>  > Would it also make sense to make struct.pack() return a buffer object
>  > instead of a string?
> 
>   Sounds like it to me.  If Guido agrees, I can make the changes.

Wait... what's so bad about buffer(struct.pack()) ? Strings already
know the buffer interface, so this works just fine already.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   147 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From paul at prescod.net  Sat Aug 28 21:39:00 1999
From: paul at prescod.net (Paul Prescod)
Date: Sun, 29 Aug 1999 01:39:00 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net>
		<slrn7sasvb.397.kzaragoza@kzaragoza.ne.mediaone.net>
		<slrn7scf7d.pa.clgonsal@keeshah.penguinpowered.com>
		<m3wvuh8yo3.fsf@pcrm.win.tue.nl>  <slrn7sf2m1.1gv.clgonsal@keeshah.penguinpowered.com>
		<slrn7sgesi.lbd.thantos@brimstone.mecha>  <slrn7sgq7j.4ee.clgonsal@keeshah.penguinpowered.com> <slrn7sgrul.sv3.thantos@brimstone.mecha>
Message-ID: <37C88F34.BBB6A94A@prescod.net>

Alexander Williams wrote:
> 
> Reducing these kinds of 'errors' isn't the function of a language.  

Python goes to great extents to reduce these sorts of errors in other
parts of the language. You'll notice that none of the old Python
programmers have argued that this change would be un-pythonic. Some of
them like the feature too much to give it up for safety but nobody
questions the goal of improving language safety. If you compare Python
to Perl you'll find that Python *is* a B&D language -- in comparison
(but is much less of one than Java). That's balance is a major virtue.

 Paul Prescod



From arcege at shore.net  Tue Aug 10 13:02:00 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 10 Aug 1999 17:02:00 GMT
Subject: Why is CGI module so complicated?
References: <37AB04A8.C3403427@NOSPAM.strs.co.uk>
Message-ID: <cWYr3.1329$EG4.226668@news.shore.net>

Ian Clarke <I.Clarke at NOSPAM.strs.co.uk> wrote:
: I have been writing an application using the cgi module, and am somewhat
: confused as to why data is returned in FieldStorage or MiniField storage
: objects, why not just return it in a plain dictionary?

: At the moment, lets say you want to access a CGI variable called "pie"
: you need to use:

: f = cgi.FieldStorage()
: print f['pie'].value

: But why not just:

: f = cgi.FieldStorage()
: print f['pie']

: That way the FieldStorage object would be a true dictionary, and could
: be used as, for example, a TemplateDocument's substitutions dictionary.

: I would write a wrapper to get rid of all of the FieldStorage objects,
: but I have almost finished my code :-(

: Can anyone explain what the point of the FieldStorage and
: MiniFieldStorage objects are?

: Ian

Ian, please read my reply in the previous thread:
  http://www.dejanews.com/getdoc.xp/AN=502365247

  -Arcege




From stephan at pcrm.win.tue.nl  Fri Aug 20 03:38:08 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 20 Aug 1999 09:38:08 +0200
Subject: A question of variables
References: <37bcf7d7.0@sahara.cablecat.com>
Message-ID: <m3u2pulx5r.fsf@pcrm.win.tue.nl>

"Carlos Ors" <cors at recercai.com> writes:

> How can I declare a variable as a private variable inside a method of a
> class?

Variables in a method are always private.
In a class, you can use names with begin with one underscore,
e.g. _foo, or with two underscores, e.g. __foo.

A name beginning with one underscore is a polite suggestion to
whoever is going to use the class in the future not to stomp
on this name. However, there is no intrinsic protection from the
language.

A name beginning with two underscores gets "mangled", i.e.
the classname gets thrown in. This means that in the case:

class X:
    __foo = 1
    def print_x(self):
        print self.__foo

class Y(X):
    __foo = 2
    def print_y(self):
        print self.__foo

,when you do subsequently

>>> inst = Y()
>>> inst.print_x()
1
>>> inst.print_y()
2

,which shows that the __foo in class X and the __foo in class Y
are actual replaced by different names, namely _X__foo and _Y__foo. 
However, the variables can still be accessed by anyone using
these mangled names!

The bottom line is that access protection in Python can always
be circumvented. However, doing so is of course a Bad Idea.

Greetings,

Stephan



From ojw at iinet.net.au  Fri Aug 20 06:55:37 1999
From: ojw at iinet.net.au (Oliver White)
Date: Fri, 20 Aug 1999 18:55:37 +0800
Subject: Setting environment variables.
Message-ID: <37BD3429.7364B4F7@iinet.net.au>

I'm writing my first python program...

The program is intended to give a menu of choices, from which the
CVSROOT environment variable is set (I'm sick of swapping around
manually for sucking down code from different places). I'm wondering how
I would do that.

I'm running GNU/Linux, is there an issue with processes... eg, after
running my python program as a child process, will the changes to the
environment variable affect it's parent (the *term)? (suspecting no) Is
there annother way to set the parent's?



From che at debian.org  Thu Aug 19 13:47:55 1999
From: che at debian.org (Ben Gertzfield)
Date: 19 Aug 1999 10:47:55 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu>
Message-ID: <ytt7lmrirw4.fsf@gilgamesh.cse.ucsc.edu>

>>>>> "Tom" == Tom Christiansen <tchrist at mox.perl.com> writes:

    Tom> At the very least, python needs a top-level python(1) manpage
    Tom> that gives calling conventions, options, envariables etc, and
    Tom> an explanation of how to locate the rest of the documentation
    Tom> (even if it's in non-man form).  That's a POSIX.2
    Tom> requirement.

I don't know about your system, but on Debian Linux 2.1, I have a
python(1) manpage, along with an entire set of HTML docs in
/usr/doc/python . The manpage includes an explanation on how to get
the Tutorial and Library Reference. Here's the first page:

PYTHON(10 April, 1998)                     PYTHON(10 April, 1998)


NAME
       python - an interpreted, interactive, object-oriented pro-
       gramming language

SYNOPSIS
       python [ -d ] [ -i ] [ -O ] [ -S ] [ -t ] [ -u ] [ -v ]  [
       -x ] [ -X ]
              [ -c command | script | - ] [ arguments ]

DESCRIPTION
       Python  is  an  interpreted,  interactive, object-oriented
       programming language that combines remarkable  power  with
       very  clear syntax.  For an introduction to programming in
       Python you are  referred  to  the  Python  Tutorial.   The
       Python  Library  Reference documents built-in and standard
       types, constants, functions  and  modules.   Finally,  the
       Python Reference Manual describes the syntax and semantics
       of the core language in (perhaps too) much detail.

Ben

-- 
Brought to you by the letters P and M and the number 6.
"What's green and sits in the corner? ... A naughty frog!"
Debian GNU/Linux maintainer of Gimp and GTK+ -- http://www.debian.org/



From jessw at loop.com  Mon Aug 30 20:39:56 1999
From: jessw at loop.com (JW)
Date: Mon, 30 Aug 1999 17:39:56 -0700
Subject: Adding manual access to IDLE
Message-ID: <37CB245C.3AF@loop.com>

I am a very new user of Python, and although IDLE is a godsend(Ooh,
colors!), it was curiously missing any access to the Python
Documentation.  I added a simple command to open the docs, but it is
pretty ugly, and I wonder if anyone else has noticed this problem or has
any sugestions about dealing with it?

						Jesse Weinstein

These are my alterations in the IDLE source:

idle\Bindings.py: 51:
    ('Open _Docs...', '<<open-docs>>'), #JW
idle\EditorWindow.py: 55:
 #$ event <<open-docs>> #JW
idle\EditorWindow.py: 112:
         text.bind("<<open-docs>>", self.open_docs)  #JW
idle\EditorWindow.py: 241:
     def open_docs(self, event=None):                #JW
     	os.system("C:\INTERNET\NETSCAPE\NAVIGA~1\PROGRAM\NETSCAPE.EXE
C:\python\doc\index.html") #JW



From gmcm at hypernet.com  Mon Aug 30 15:21:25 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 30 Aug 1999 14:21:25 -0500
Subject: 'if foo' calls getattr?!?
In-Reply-To: <joe-300819991050234291@chinacat.salk.edu>
Message-ID: <1276085477-52750363@hypernet.com>

Joe Strout wrote:
 
> Doing some profiling on Graphite this weekend, we found a very
> curious thing: a gazillion calls to __getattr__ on one of our
> classes, where we expected to find none. 
> 
> Below is a very simple stand-alone demonstration.  It leads me to
> the absurd idiom of saying "if id(foo) != id(None)" when I really
> mean "if foo", just because using id() avoids the __getattr__ call.
> 
> Can anybody explain why this should be?  Is it a bug in Python? 
> 
> getattrcount = 0
> 
> class Foo:
>    def __getattr__(self,attr):
>       global getattrcount
>       getattrcount = getattrcount+1
>       return self.__dict__[attr]
> 
> def testA():
>    global getattrcount
>    getattrcount = 0
>    foo = Foo()
>    for i in range(42):
>       if foo: pass
>    print "calls due to 'if foo':", getattrcount

If you read the Timbot's posts, you'd know it's not the __getattr__, 
it's the "42"!

More instructive is to see what "attr" is: First it's __nonzero__, 
and then it's __len__. Python is doing whatever it can to determine 
whether your object considers itself "true" or not. Which leads to 
the conclusion that "if X" doesn't test if X exists, but whether X is 
"true" according to some interpretation.

You've found one solution. You don't say what the alternative to foo 
is; if that alternative is non-existance, you've got a NameError.  So 
I'll assume the alternative is None.

In which case "if foo is not None" will be the fastest and 
closest-to-your-intention test possible. (But "if foo != None" will 
get you right back to where you are now!).

- Gordon



From laheadle at boguscs.uchicago.edu  Sun Aug 22 20:19:58 1999
From: laheadle at boguscs.uchicago.edu (Lyn A Headley)
Date: Mon, 23 Aug 1999 00:19:58 GMT
Subject: matching any number of any character
Message-ID: <wr34shrtkk1.fsf@gargoyle.cs.uchicago.edu>

To do it, I want to use the regex module, heretical though that may
be.  I think my brain is missing a piece, because I can't seem to
accomplish such a task.  I thought this would work:

>>> import regex
>>> rx = regex.compile('\(.\|\n\)+')
>>> rx.match('abc')
3
>>  rx.group(1)
'c'

but I wanted 'abc'!

little help?

-Lyn



From JamesL at Lugoj.Com  Sun Aug 15 17:31:17 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Sun, 15 Aug 1999 14:31:17 -0700
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net> <slrn7re808.77t.kaz@ashi.FootPrints.net>
Message-ID: <37B731A5.13986DA7@Lugoj.Com>

Kaz Kylheku wrote:
> 
> On 15 Aug 1999 03:14:48 GMT, Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
> >[strange programming]
> >The most confidence uninspiring piece of code I ever had the displeasure to
> >maintain contained something like this (allegedly C++, but mainly it's
> >intersection with C):
> >
> >       i = 0;
> >       while( i < max_index ) {
> >               /* some copying or update (not modifying i) goes here */
> >               i = i + 1;
> >       }
> >
> >This was written by a *very* expensive consultant working for a large
> >multinational accountancy firm.
> 
> So what's wrong with that?
[Nice defense elided.]

Those that think that code is bad have I suspect have never seen bad code.
Anyway, I bet many people missed the really bad part in that code, which is
the placement of the { <Wink>. Note the only approved location for curly
brackets:

i = 0;
while(i < max_index)
{
    /* Blah blah. */
    i = i + 1;
}

(I use Python whenever I can just to get away from these curly bracket wars.
<0.7 wink>)



From robin at alldunn.com  Mon Aug  9 15:53:57 1999
From: robin at alldunn.com (Robin Dunn)
Date: Mon, 9 Aug 1999 12:53:57 -0700
Subject: ANNOUNCE:  wxPython 2.1b2
Message-ID: <024c01bee2a0$ec84bf40$1a25d2d1@jenkondev.com>

wxPython 2.1b2

wxPython is an extension module for Python that wraps the wxWindows cross
platform GUI library, and is quickly becoming a very popular alternative to
Tkinter and PythonWin. This extension module attempts to mirror the class
hierarchy of wxWindows as closely as possible and is very versatile. It can
be used to create standalone GUI applications, or can be used in in
situations where Python is embedded in a wxWindows C++ application as an
internal scripting or macro language. Currently supported GUIs are Win32 and
GTK/X-Windows.

Changes in this version:

Added the missing wxWindow.GetUpdateRegion() method.

Made a new change in SWIG that provides a fix for global shadow
objects that get an exception in their __del__ when their
extension module has already been deleted. It was only a 1 line
change in .../SWIG/Modules/pycpp.cxx at about line 496 if you want
to do it by hand.

It is now possible to run through MainLoop more than once in any one
process.  The cleanup that used to happen as MainLoop completed (and
prevented it from running again) has been delayed until the wxc module
is being unloaded by Python.

I fixed a bunch of stuff in the C++ version of wxGrid so it wouldn't
make wxPython look bad.

wxWindow.PopupMenu() now takes a wxPoint instead of  x,y.  Added
wxWindow.PopupMenuXY to be consistent with some other methods.

Added wxGrid.SetEditInPlace and wxGrid.GetEditInPlace.

You can now provide your own app.MainLoop method.  See
wxPython/demo/demoMainLoop.py for an example and some explaination.

Got the in-place-edit for the wxTreeCtrl fixed and added some demo
code to show how to use it.

Put the wxIcon constructor back in for GTK as it now has one that
matches MSW's.

Added wxGrid.GetCells

Added wxSystemSettings static methods as functions with names like
wxSystemSettings_GetSystemColour.

Removed wxPyMenu since using menu callbacks have been depreciated in
wxWindows.  Use wxMenu and events instead.

Added alternate wxBitmap constructor (for MSW only) as
      wxBitmapFromData(data, type, width, height, depth = 1)

Added a helper function named wxPyTypeCast that can convert shadow
objects of one type into shadow objects of another type.  (Like doing
a down-cast.)  See the implementation in wx.py for some docs.

Fixed wxImage GetData and SetData to properly use String objects for
data transfer.

Added access methods to wxGridEvent.

New Makefile/Setup files supporting multiple dynamic extension modules
for unix systems.

Fixes for the wxGLCanvas demo to work around a strange bug in gtk.

SWIG support routines now compiled separately instead of being bundled
in wx.cpp.




--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!





From tdfunk at asd-web.com  Thu Aug 26 17:59:57 1999
From: tdfunk at asd-web.com (Tom Funk)
Date: Thu, 26 Aug 1999 17:59:57 -0400
Subject: Python article on Builder.com
Message-ID: <7DB247B399FED211A73F00600841DDC7099723@TAL_ENT>

Thanks, John:

If you haven't aleady, do you mind if I post your comments to the article's
discussion area at builder.com?

-=< tom >=-

-----Original Message-----
From: jonathanbrisbin at excite.com [mailto:jonathanbrisbin at excite.com]
Sent: Thursday, August 26, 1999 4:46 PM
To: Tom Funk; python-list at python.org
Subject: Re: Python article on Builder.com


On Thu, 26 Aug 1999 14:34:11 -0400, Tom Funk wrote:

> Some unanswered issues have been raised that could use some input from 
> some experienced Python developers.  Also, some comments from someone
> who's used both Perl and Python would be welcome (only one such reader
> has chimed in to date).
> 

Well, I'm in the process of replacing a number of my perl scripts with
python because I like programming in Python better than in perl, the .pyc's
load noticeably quicker, and it has some great features (threads!) that make
it very scalable.  In fact, I can create a self-contained webserver and
content delivery system entirely in python (a project I'm actually working
on :-).  As far as database access, I wrote a perl script and a python
script that turned out to be exactly the same number of lines and fetched
stuff from a MySQL database.  I'm running mod_perl and mod_pyapache in my
Apache 1.3.6 and the python reloaded noticeably quicker.  Throw threads into
the mix and things start working pretty quick...

Just some thoughts...

><> <>< ><> <>< ><> <><

Jon Brisbin
  JonathanBrisbin at excite.com




From fdrake at acm.org  Thu Aug 26 10:00:06 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 26 Aug 1999 14:00:06 GMT
Subject: C module leaking memory
In-Reply-To: <37C4DC02.32ADCC3A@inka.de>
References: <FH1wtI.C68@world.std.com>
	<37C4DC02.32ADCC3A@inka.de>
Message-ID: <14277.18534.707277.428011@weyr.cnri.reston.va.us>

Friedrich Dominicus writes:
 > One of the first points you can try is -Wall +W extra-options (for gcc )
 > and using lclint 

  Do you (or does anyone else) have an lclint interface description
for Python?  I was interested in using it, but don't have the time to
develop the formal description of the Python C API.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From angela at objectzoo.com  Mon Aug 16 03:40:17 1999
From: angela at objectzoo.com (Rudi Angela)
Date: Mon, 16 Aug 1999 09:40:17 +0200
Subject: Python CGI Scripts
References: <Pine.LNX.4.04.9908150140180.584-100000@gnr.u2me3.com>
Message-ID: <37B7C061.3824206C@objectzoo.com>

Stephen wrote:
> 
> But... have a lot of people adopted python as their new cgi language? This is
> what I do primarily lately and I found only one program in the book called
> 'cgimail.py'. As I read through the code I notice that the h1 tags, the form,
> the header infomation, etc etc are all being 'print'ed manually. Is there no
> nifty CGI.pm equivelent in python?
> 
> Comments?
> 

Yep, have a look at Zope, a python based opensource cgi solution:

http://www.zope.org/

Cheers,

Rudi Angela



From jstevens at basho.fc.hp.com  Fri Aug 13 23:08:00 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:08:00 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132020.OAA03699@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:20 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cf80@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :But in a test where programmers are ignorant of each, Python is
> :the prefered choice . . . *EXCEPT* where the programmers were
> :experienced C programmers.
> 
> Well what else are you going to get?  BASIC programmers?  Now those
> are clever folks for you!  Of course most programmers are C (or C++)
> programmers.

Sigh.  Not "of course".  There are quite a few programmers, now, who
don't have a C/Unix background.

And, though I failed to mention it (I was trying to be a little bit
light), many of those C programmers came around after a while.

> It's the only real-world langauge that gets the jobs
> done in millions of environments.  Every CompSci student should be
> proficient at C/C++.  Anything else is fluff.  Beneficial, but fluff.

Wow.  Strong words.

I note, with some sadness, that you failed to mention Fortran or
Cobol, languages that also get the job done, in millions of
real world environments.  Fortran and Cobol are fluff, hunh?

> I don't care whether they're proficient at six other languages, and
> in fact, I hope they know those six others, but C/C++ is what counts.
> What kind of "programmers" are you talking about?

Real programmers.  Those who do it for a living.  Some do C, others do
C++, yet others use Cobol, Fortran, Visual Basic, Objective-C, assembler,
Lisp, Prolog, Pascal, etc.

The whole world is *NOT* C.

> I think your newsreader is buggy.  Better fix it.

My news reader works fine.  Your news server or your news reader are
flawed.  Better fix 'em. . .

John S.




From fredrik at pythonware.com  Fri Aug 20 09:38:53 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 13:38:53 GMT
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk> <37BC8D05.142E1DB0@aw.sgi.com> <37BC9CE8.67097FC6@vision.arc.nasa.gov> <37BCD6CF.B7A186BC@montana.com> <5lzozmskm3.fsf@eric.cnri.reston.va.us>
Message-ID: <00dd01beeb11$e41a68e0$f29b12c2@secret.pythonware.com>

Guido van Rossum <guido at cnri.reston.va.us> wrote:
> Actually, the accepted idiom (all over the place in code I write) is
> 
>         toplevel_window.protocol("WM_DELETE_WINDOW", close_callback)
> 
> The close_callback function takes no arguments.  In simple cases, you
> can use
> 
> root.protocol("WM_DELETE_WINDOW", root.destroy)
> 
> or
> 
> root.protocol("WM_DELETE_WINDOW", root.quit)

how about making the destroy variant the
default?

</F>




From bwarsaw at cnri.reston.va.us  Fri Aug 20 11:40:42 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Fri, 20 Aug 1999 11:40:42 -0400 (EDT)
Subject: making python man pages looks hard
References: <1103_935134343@shoal>
Message-ID: <14269.30458.285805.329231@anthem.cnri.reston.va.us>

>>>>> "DC" == Dan Connolly <connolly at w3.org> writes:

    DC> The bewildering array of scripts, tools, and hacks used to
    DC> generate the HTML version of the python docs is frightening!

I think that's the Quote of the Day.  Wouldn't you agree Fred? :)

    DC> It suggests to me that *very few people* maintain the
    DC> python docs.

few == one.  Fred Drake, a.k.a. Dr. Doc.

speaking-as-a-beta-tester-of-fred's-html-documentation-ly y'rs,
-Barry



From jfisher at are.berkeley.edu  Mon Aug 23 16:27:34 1999
From: jfisher at are.berkeley.edu (John Fisher)
Date: Mon, 23 Aug 1999 20:27:34 GMT
Subject: gcc optimization breaks NumPy?
Message-ID: <7psar7$kt1$1@nnrp1.deja.com>

Hey folks,

I was getting consistent but inexplicable segmentation faults with the
following code, using NumPy's C API to move data between Python and C
(for sparse matrix multiplication; the data structure is from Meschach,
in case that's relevant).  I found the problem to be caused by gcc's
optimization options.

static PyObject * mul(PyObject *self, PyObject *args) {

  ...Snip declarations...

  ...Snip working code.  Function ends with:

  return MakeFromMes(product);
}

which is defined as:

PyObject * MakeFromMes(SPMAT *in) {
  int i, j, m, n, nzs, count, pos, therow, thecol;
  int dummy[0];
  double elem;
  PyArrayObject *pr, *ir, *jc;

  m = in->m; n = in->n; nzs = 0;

  /* Count the number of nonzero elements of in
     ISZERO is a macro to test for "good enough" floating point 0 */
  for(i = 0; i < m; i++) {
    for(j = 0; j < n; j++) {
      if(!ISZERO(sp_get_val(in, i, j))) nzs++;
    }
  }

  dummy[0] = nzs;

  if(!( (pr = PyArray_FromDims(1, dummy, PyArray_DOUBLE)) &&
	(ir = PyArray_FromDims(1, dummy, PyArray_INT)) &&
	(jc = PyArray_FromDims(1, dummy, PyArray_INT)) )) {
    fprintf(stderr, "MakeFromMes: Could not create output arrays.\n");
    return Py_BuildValue("O", Py_None);
  }

  count = pos = 0;

  /* Fill column-wise */
  for(j = 0; j < n; j++) {
    for(i = 0; i < n; i++) {
      elem = sp_get_val(in, i, j);
      if(!ISZERO(elem)) {
	therow = pos % n;
	thecol = (int)(pos / n);
	*(double *)(pr->data + count * pr->strides[0]) = elem;
	*(int *)(ir->data + count * ir->strides[0]) = therow;
	*(int *)(jc->data + count * jc->strides[0]) = thecol;
	count++;
      }
      pos++;
    }
  }
  return Py_BuildValue("OOO", pr, ir, jc);
}

The segmentation fault occured immediately after MakeMes() returned.
Were I to change the mul() function to end with

  PyObject *out = MakeFromMes(prod);
  printf("Hello.");
  return out;
}

the error would occur even before execution passed to the printf.

The problem occurs only with an -O option in compilation.  Other
specific -f optimization options do not cause the seg fault.  So my
question is -- because I've let gcc optimize other extensions to Python
I've written without incident -- what was it here that caused this
behavior?  I'd appreciate any ideas and speculations, because I'm still
baffled by this, though the problem seems to be solved.

Thanks,
John


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From posselt at daimi.au.dk  Mon Aug 23 12:01:15 1999
From: posselt at daimi.au.dk (Peter Posselt Vestergaard)
Date: Mon, 23 Aug 1999 18:01:15 +0200
Subject: Sending a mail
Message-ID: <37C1704B.D1B173E5@daimi.au.dk>

Hi.
I want my script to send a mail with a specific subject. I've tried to
import and use the smtplib, but I get the message that it doesn't exist
- maybe it isn't implemented on the server I'm using, and anyway - I
can't specify a subject using the sendmail()-command...
Then I tried to use the os.system like follows:

#!/usr/local/bin/python
import os
text='echo "hello"|mailto posselt at daimi.au.dk -s "important message"'
os.system(text)

This works quite fine in my python interpreter, but not then I put it on
the web. I tried to print out the standardoutput from os.system(text)
and it was 256. Probably my python-server won't allow me to execute that
command or what? I hope that anyone knows some kind of workaround, or
another way to do this???
Best regards
Peter Posselt Vestergaard



From skip at mojam.com  Sat Aug 21 20:50:51 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sat, 21 Aug 1999 19:50:51 -0500 (CDT)
Subject: Extending Python: exception handling
In-Reply-To: <37bdcb30.796706964@news.triode.net.au>
References: <37bdcb30.796706964@news.triode.net.au>
Message-ID: <14271.18469.757616.424717@dolphin.mojam.com>

    John> I would like to know how YOU are using exception handling OTHER
    John> than for error management.

It makes a reasonable substitute for the lack of labelled break statements
on certain occasions.

    class BreakLabel: pass
    try:
	for i in ...:
	    for j in ...:
		exit_cond = do_some_work()
		if exit_cond: raise BreakLabel
    except BreakLabel:
	pass

instead of 

    for i in ...:
	exit_cond2 = 0
	for j in ...:
	    exit_cond = do_some_work()
	    if exit_cond:
		exit_cond2 = 1
		break
        if exit_cond2: break

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From Sven.Drescher at dlr.de  Fri Aug 20 06:01:59 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Fri, 20 Aug 1999 12:01:59 +0200
Subject: Embedding python
References: <7pdnso$om6@news.go.dlr.de>
Message-ID: <7pj88p$pk2@news.go.dlr.de>

Hi!

I have found a very mad solution.

Once I used an other compiler and just all is ok.
Gcc is just the best!!!

Thanks!

Bye,
Sven






From I.Clarke at strs.co.uk  Fri Aug 20 05:06:18 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Fri, 20 Aug 1999 10:06:18 +0100
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu> <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <37BD1A8A.D4DCFC7C@strs.co.uk>

> Ah, have you discovered rlcompleter yet? Put this in your ~/.pythonrc:
> 
> import rlcompleter
> readline.parse_and_bind('tab: complete')
> 
> set PYTHONSTARTUP=~/.pythonrc and then in python type:
> 
> d={}

Is there any way to get this working on Windoze?

Ian.



From jeremy at cnri.reston.va.us  Thu Aug 19 14:24:43 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 19 Aug 1999 14:24:43 -0400 (EDT)
Subject: How to sell Python to your boss?
In-Reply-To: <11A17AA2B9EAD111BCEA00A0C9B41793034AAD03@molach.origin.ea.com>
References: <11A17AA2B9EAD111BCEA00A0C9B41793034AAD03@molach.origin.ea.com>
Message-ID: <14268.19435.271581.609@bitdiddle.cnri.reston.va.us>

I think you need to take David's message to your boss with the
following introduction:

Many companies that are doing cutting edge work in this field use
Python -- companies like Electronic Arts and Industrial Light and
Magic.  These are big companies with talented programmers and money to
invest in the right tools.  Your project will benefit from Python for
the same reasons that these companies projects benefit from Python.
You might also pick up some skills that would help you find a job.

Jeremy




From fdrake at acm.org  Thu Aug  5 11:15:41 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 5 Aug 1999 15:15:41 GMT
Subject: [Image-SIG] Proper application of the buffer interface
In-Reply-To: <Pine.WNT.4.05.9908042019270.40-100000@david.ski.org>
References: <Pine.WNT.4.05.9908042019270.40-100000@david.ski.org>
Message-ID: <14249.43677.175790.689266@weyr.cnri.reston.va.us>

David Ascher writes:
 > Would it also make sense to make struct.pack() return a buffer object
 > instead of a string?

  Sounds like it to me.  If Guido agrees, I can make the changes.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From MHammond at skippinet.com.au  Thu Aug 26 00:50:22 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 26 Aug 1999 14:50:22 +1000
Subject: OCX and PythonWin....
References: <7oq27u$th0$1@nnrp1.deja.com> <%oat3.62670$jl.39620751@newscontent-01.sprint.ca>
Message-ID: <7q2hi1$6qb$1@m2.c2.telstra-mm.net.au>

[cc'd to poster]

Bruce Dodson wrote in message
>On a similar subject.  I was experimenting with MapObjects (an ActiveX
>control) and PythonWin.  It seems that every time I load it, PythonWin
>crashes on exit (access violation or something).  I tried generating a
quick
>AppWizard test in VC++, and had no such problems there.  Also no problems
in
>VB, of course.

This is almost certainly due to one or both of:
* The COM object is still alive.  pythoncom._GetInterfaceCount() must return
zero before you can safely exit.
* You need to call pythoncom.CoUninitialize()


>There is some benefit to this pain.  I am slowly coming to grips with the
>way PythonWin wraps MFC.  A light turned on as I was writing this and
>realized why the docs talk about PyCMDIChildWnd while the examples derive
>from things like MDIChildWnd, why the latter isn't mentioned in the help
>file whereas the former can't be found in the object browser.  But it's
>6:00am.  After a good morning's sleep I will probably have forgotten what I
>learned. :-)

This is tricky and not obvious.  The PyCMDIChildWnd is a Python _type_,
implemented in win32ui.pyd.  Unfortunately you can not derive from a Python
type, only a Python class.  So a framework has been put in place that allows
you to use the Python class (eg pywin.mfc.window.MDIChildWnd) as if it were
the PyCMDIChildWnd type.

This is vaguely documented somewhere, but I can't recall where and there are
still 1000 unread messages in this newsgroup for me to get through :-)

Mark.






From jblaine at shore.net  Tue Aug 17 09:39:30 1999
From: jblaine at shore.net (Jeff Blaine)
Date: Tue, 17 Aug 1999 13:39:30 GMT
Subject: SUMMARY: C's isprint() concept
Message-ID: <mCdu3.363$mV.139731@news.shore.net>

Thanks to those of you who replied with help or attempted help and no
thanks to those of you who replied and were snippy.

Method 1:  Loop over every character in the line, check ord() value

Method 2:  Use a regexp (re.sub)

Method 3:  Use string.maketrans and string.translate (Python dist
           documentation on these methods needs work if you ask me).

Results:   Method 3 is fastest by far.  Method 1 was second fastest.  The
           regexp method was slowest.

#-----------------------------------------------------------------------------
# Snippet from Method 1 - assumes an open fd
#-----------------------------------------------------------------------------
for line in fd.readlines():
    good = []
    for character in line:
        ordval = ord(character)
        if ordval < 32 or ordval > 126:
            good.append(' ')
            continue
        else:
            good.append(character)

#-----------------------------------------------------------------------------
# Snippet from Method 2 - assumes an open fd
#-----------------------------------------------------------------------------
NOTPRINTABLE = r'[^ -~]+'
npre = re.compile(NOTPRINTABLE)
for line in fd.readlines():
    new_line = re.sub(npre, ' ', line)

#-----------------------------------------------------------------------------
# Snippet from Method 3 - assumes an open fd
#-----------------------------------------------------------------------------
ttable = string.maketrans('\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1A\x1B\x1C\x1D\x1E\x1F\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8A\x8B\x8C\x8D\x8E\x8F\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9A\x9B\x9C\x9D\x9E\x9F', '                                                                ')
for line in fd.readlines():
    new_line = string.translate(line, ttable)



From bradh at mediaone.net  Thu Aug 19 21:22:33 1999
From: bradh at mediaone.net (Brad Howes)
Date: 19 Aug 1999 21:22:33 -0400
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <8E27B3FC4duncanrcpcouk@news.rmplc.co.uk>
Message-ID: <wjyu2pv2qli.fsf@bradh.ne.mediaone.net>

duncan at rcp.co.uk (Duncan Booth) writes:

> Or if you know you are going to have a tuple, you could declare the 
> function to take a tuple as its argument:
> def fn((x,y)): ...

Get out! You can do that?! Wow -- learn something new every day. Thanks!

Brad

-- 
Brad Howes                    bradh at mediaone.net



From tim_one at email.msn.com  Fri Aug 20 01:19:00 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 20 Aug 1999 01:19:00 -0400
Subject: sprintf & round (was RE: More random python observations from a perl programmer)
In-Reply-To: <37BC1ED4.A29883BC@appliedbiometrics.com>
Message-ID: <000901beeacb$83212240$712d2399@tim>

[Tom Christiansen]
>     Python has a round() function, but it seems to ignore the
>     IEEE rules of rounding.  However, its sprintf operator doesn't:
>         >>> print "int of %f is %.0f" % (1.5, 1.5)
>         int of 1.500000 is 2
>         >>> print round(1.5,0)
>         2.0
>         >>> print "int of %f is %.0f" % (2.5, 2.5)
>         int of 2.500000 is 2
>         >>> print round(2.5,0)
>         3.0

Python's sprintf rounding is inherited from the platform libc's.  Virtually
all systems will round 1.5 up to 2.0, but there's wide variation (even among
Unices) in what %f.0 does to 2.5.  The "round" function is Python's own.

[while Christian Tismer gets ...]
> I had a look at this again. Despite the fact that the text in
> the above example is a bit misleading by "int of" where
> a round() takes place, which version of Python/architecture
> did you use? I get the expected result, see below.
>
> >>> print "round of %f is %.0f" % (1.5, 1.5)
> round of 1.500000 is 2
> >>> print round(1.5,0)
> 2.0
> >>> print "round of %f is %.0f" % (2.5, 2.5)
> round of 2.500000 is 3
> >>> print round(2.5,0)
> 3.0
> >>>

My guess is you're running on Windows.  MS rounds all halfway cases up;
Tom's complaint is that under IEEE-754 to-nearest/even rules, 1.5 and 2.5
should both round to 2.  Luckily, Python's will do that the instant MS's
does <wink>.

not-sure-there's-a-winner-in-arguing-over-which-language-has-the-
    lamest-754-support-ly y'rs  - tim





From paul at prescod.net  Tue Aug 10 22:57:01 1999
From: paul at prescod.net (Paul Prescod)
Date: Wed, 11 Aug 1999 02:57:01 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com>
		<37B02965.B1FE0E8D@prescod.net> <934326002snz@vision25.demon.co.uk>
Message-ID: <37B0E67D.27F8D2ED@prescod.net>

Phil Hunt wrote:
> 
> In article <37B02965.B1FE0E8D at prescod.net>
>            paul at prescod.net "Paul Prescod" writes:
> > Have you looked into XUL,
> 
> Glade is a GUI application that allows you to build a window visually.

Okay, but I saw a reference somewhere to Glade having a declarative file
format that could be converted into code to build the GUI. I was mostly
talking about the declarative file format.

> My system (provisional name: ``Parrot'') is a command-line application
> that takes an ascii file as input, and outputs a file containing source
> code in your favourite language, to build the GUI.

Glade does that too, as long as your "favorite language" is C. :)

I don't understand why you would take a code generation approach,
though. My professional opinion is that code generation is yucky. :) It
takes no more effort to write a "Parrot reader" for each language --
especially if Parrot uses the language's built-in XML support (oops.
scratch that). Seriously though, XML aside, why bother with code
generation when you can just do everything dynamically? Read the parrot
file and make the objects on the screen. The performance shouldn't be
much different and the extra dynamicity will make your life easier and
make the tool cooler to use. "I'll just generate a modified parrot file
and reload."

> >  * it can be trivially parsed with existing code written in several
> > languages
> 
> True; I don't think a parser for Python will be too difficult to
> write: my provisional spec only has 12 productions.

These specs always start out small. :)

> True, but writing classes implementing recursive tree-structures
> is almost trivial in Python. (One reason I like it).

Wait until you get to the TCL version!

> What's XUL?

Netscape's version of the same thing you are doing:

http://www.oasis-open.org/cover/xul.html

> I will stick with my current format; but adding a feature to
> import/export to XML shouldn't be too difficult -- then I'd have
> the best of both worlds.

True enough. I think history shows that one syntax or another always
ends up dominating, though. Lisp had two syntaxes once...

 Paul Prescod



From hoel at GermanLloyd.org  Tue Aug 31 10:32:00 1999
From: hoel at GermanLloyd.org (Berthold Hoellmann)
Date: Tue, 31 Aug 1999 16:32:00 +0200
Subject: problem with execfile
Message-ID: <37CBE760.3227B3A0@GermanLloyd.org>

Hello,

I have a problem with execfile:

python
Python 1.5.2 (#2, Jul  7 1999, 09:18:40) [C] on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> def x():
...     execfile("ZustandsDaten.dat")
...     print dir()
...     print Tiefgang
... 
>>> x()
['Pitch', 'Tiefgang', 'U', '__doc__', 'array', 'n']
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "<stdin>", line 4, in x
NameError: Tiefgang
>>> 

Why is Tiefgang unknown in the function x?

Greetings 

Berthold
-- 
email: hoel at GermanLloyd.org
   )
  (
C[_]  These opinions might be mine, but never those of my employer.



From fdrake at acm.org  Mon Aug 23 15:24:04 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Mon, 23 Aug 1999 19:24:04 GMT
Subject: More random python observations from a perl programmer
In-Reply-To: <093501beeb7b$c37ff400$0301a8c0@mnot.net>
References: <199908191634.KAA01333@jhereg.perl.com>
	<37BC75F0.B482DA6D@starmedia.net>
	<37bc4352@cs.colorado.edu>
	<37BCA74C.D3DCF5DD@fnmail.com>
	<3bgv3.759$mV.257016@news.shore.net>
	<093501beeb7b$c37ff400$0301a8c0@mnot.net>
Message-ID: <14273.40916.612267.803684@weyr.cnri.reston.va.us>

Mark Nottingham writes:
 > Not as a perldoc-ish sort of thing, but IMHO it would be helpful to have a
 > standard of documentation and installation for extention modules (as well as
 > library add-ons).

Mark,
  The Distutils-SIG is working on module installation; see
http://www.python.org/sigs/distutils-sig/ for information about
subscribing to the mailing list.
  The Doc-SIG is the forum for discussing documentation issues for
Python; I welcome constructive criticism, suggestions, and proposals
on that mailing list.  See http://www.python.org/sigs/doc-sig/ for
mailing list information.
  Bug reports pertaining to the documentation ("XXX is wrong", "XXX is 
missing" where XXX is specific) should go to python-docs at python.org.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From mithy at my-deja.com  Thu Aug  5 13:20:26 1999
From: mithy at my-deja.com (Preston Landers)
Date: Thu, 05 Aug 1999 17:20:26 GMT
Subject: downloading web based email
Message-ID: <7och4d$jr0$1@nnrp1.deja.com>

 Hello everyone....

I'm thinking of writing a Python program to download email from
web-based services such as Yahoo mail, Deja, Hotmail, and so on, and
deliever the mail in a format that standard unix mailers can use.

I'm posting here to make sure that this work has not already been done.
If someone has already seen a program like this, that can be adapted for
a variety of web based email services, please let me know.  I'm not
writing this as a learning exercise,  I just need the tool. ;-)

My idea was to have a nice two-level effect with a path running down the
middle... no wait, I meant, a two-level effect with the higher level
handling the actual delievery of the mail and a variety of modules to
handle the quirks of each particular mail service.

If I get something coded up, I'll be sure to post here about it.  Just
wondering if somebody's already attempted this.

thanks,

---Preston

--
||  Preston Landers <mithy at my-deja.com>  ||
||        http://askpreston.com          ||


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From Sven.Drescher at dlr.de  Mon Aug  2 05:14:33 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Mon, 2 Aug 1999 11:14:33 +0200
Subject: Libraries needed, please help!!!
References: <7ns6o5$jo2@news.go.dlr.de> <004e01beda98$d14f9390$f29b12c2@secret.pythonware.com>
Message-ID: <7o3moo$dse@news.go.dlr.de>

Hi!

I did it.
The only file I found was 'libpython1.5.a'.
This you told me in the other mail.

Ok. Thanks!

But how can I get the other files (libModules.a, libPython.a, libParser.a,
libObjects.a) from libpython1.5.a?

Or can anybody send me these files (SUN/Solaris OS)???
All trials with make and configure were not successful.

Thanks for more hints and comments!

Sven





From cgw at fnal.gov  Tue Aug 17 18:38:57 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 17 Aug 1999 22:38:57 GMT
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<14265.47748.241386.291144@buffalo.fnal.gov>
	<wkpv0mgna5.fsf@turangalila.harmonixmusic.com>
Message-ID: <14265.58497.578663.958576@buffalo.fnal.gov>

Dan Schmidt writes:
 > 
 > Skip's method is fine if choice is allowed to equal i.  With that
 > change, element exchange is no longer forced.  You can prove that
 > it's a perfect shuffle by induction: the last element in the shuffled
 > sequence has an equal chance of being any of the original elements,
 > and so on.
 > 

Correct.  However, I'm curious, what is the definition of a "perfect
shuffle"?  Just that each element of the permuted sequence has an
equal chance of being any of the original elements?  As I tried to
point out in my earlier posting, there are plenty of ways of choosing
permutations for which this condition holds, but still don't uniformly
sample the symmetric group (set of all permutations).  For instance,
if you cyclically permute a list by a random amount, the above
equidistribution property holds, but you wouldn't say the sequence is
well-randomized.  There must be stronger criteria.  I almost think you
have to look at the expectation values of all the characters
corresponding to all of the irreducible representations of Sym(N).  I
looked at this for the non-trivial 1-dimensional representation given
by 'sign', but even that is not enough - you could have degenerate
methods of sampling from Sym(N) which gave a mean of 0 on this
character but had some bias on one of the higher-dimensional
characters - such an example might be a little hard to construct,
though.

Another, more straightforward way to test sampling strategies is just
to keep track of how many times each permutation comes up.  However,
since N! gets so large, it takes too long to build up decent
statistics this way.










From amaranda at nospam.com  Sat Aug 14 07:06:11 1999
From: amaranda at nospam.com (Alex Maranda)
Date: Sat, 14 Aug 1999 12:06:11 +0100
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au>
Message-ID: <37B54DA3.4210DC10@nospam.com>

Sam Holden wrote: 
> >
> >>You can write Perl programs that resemble sed, or awk, or C, or Lisp, or
> >>Python. This is Officially Okay in Perl culture.
> >
> >You cannot write Perl that resemble Python.  You are required to
> >use curly braces as block delimiters.
> 
> In fact you are not. You are wrong yet again. Actually thinking about how
> perl works for a minute would show you that perl is in fact flexible.
> it is not a lie when people say perl doesn't force you to write a
> particular way.
> 
> Here is some code from Damian Conway from the 'Impythonating PERL' thread
> in march.
> 
> package impythonate;
> use Text::Tabs;
> my ($active, @bracket) = (0, ('{', ';', '}') );
> sub import
> {
>   return 1 if ($active++);
>   open CODE, $0 or die "couldn't translate $0";
>   my ($code, $transcode, $lastindent) = (join('',<CODE>), '');
>   $code =~ s/\\\n/ /g;
>   foreach (split /\n/, $code)
>   {
>   my ($indent,$notempty) = /\A([ \t]*)(\S?)/;
>   $indent = length(expand($indent));
>   $transcode .= $bracket[1+($lastindent<=>$indent)]
>     if $notempty && defined $lastindent;
>   $transcode .= "$_\n";
>   $lastindent = $indent if $notempty;
>   }
>   eval $transcode and exit or die $@;
> }
> 1;
> 
> And here is his sample code that is now valid perl (although anyone who uses
> it for real code should be killed) :
Why? You don't happen to hate Pythoneers or Python itself, do you? I've
actually appreciated the argument from your second post, when you taxed
the other guy for needlessly questioning your Python knowledge out of
context.
I am completely ignorant with Perl and I can barely read what's above
(by interpolating my knowledge of shell, C, and Tcl).

> 
> #Now newlines replace colons and indentation replaces brackets:
> 
> use impythonate; # STILL NEED THAT ONE SEMICOLON, DAMMIT!
> 
> for $i (1..10)  # COMMENTS ARE OKAY
>   print "$i: "
>   my $isq =  \
>   $i **2 # LINE CONTINUATIONS WORK AS IN PYTHON
>   print " $isq\n"
> 
> print "done\n"
I know Python fairly well (but I'm no expert) and I can easily read this
piece of Perl. Getting back to the killing part, which kind of disturbs
me when reading comp.lang.python:
- was that necessary to make a point?
- if yes, what was the point?
- if the point (just guessing) was that Perl code should never look like
Python, isn't that against of 'there should be more ways of doing it?'

I fail to see the interesting part in this whole thread, which disturbs
the habitat of comp.lang.python. I think John Stevens was mistaken to
bring the discussion over to this newsgroup in the first place. We don't
have strong feelings about Perl, and don't have plans for world
domination or anything.

A couple of days ago I read a bit in Bugzilla's README which amused me:
"A computer which doesn't have Perl installed is a very sad computer
indeed". Your post is frightening (unless it was a Perl-style joke).

Regards (I guess),
Alex



From tim_one at email.msn.com  Fri Aug 27 00:06:19 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 27 Aug 1999 00:06:19 -0400
Subject: A Plea For A Return To Sanity!!!
In-Reply-To: <37C5659C.8892E52@callware.com>
Message-ID: <000001bef041$84deffa0$592d153f@tim>

[Ivan Van Laningham]
> 1)  If the Perl yahoos would quit cross-posting flaming bags of shit,
> it'd be over.

Actually, I'm afraid my reply here (as well as yours) keeps it going.

> 2)  Now that Tom C has left for more fertile pastures, we have been
> visited with Mr. X.  I wonder if we have 5 more visitations to go?

If they're as funny as Mr. X, we can only hope.  You should read Xah's posts
again!  They're a good parody of the worst of "Usenet Attitude" postings.  We
may have a potential Python convert there <wink>.

> 3)  The real reason for this newsgroup/mailing list is, ipso facto,
> the discussion of Python.  I don't care for having to put on my
> asbestos bunny suit just to read my mail.

Well, I read all of 'em, and they really were mostly about Python.  Besides,
Tom's very public misunderstandings are a *great* base on which to build a
"Python for Perlers" FAQ.  Python isn't self-evident at first glance, and it's
good to relive the initial confusions through his articulate eyes <ahem> (you
may not have liked how he phrased things, but there was no doubt about what he
was saying -- that's good writing).

or-is-this-a-competition-to-see-who-has-the-thinnest-skin?-ly y'rs  - tim





From abigail at delanet.com  Sun Aug 15 09:38:12 1999
From: abigail at delanet.com (Abigail)
Date: 15 Aug 1999 08:38:12 -0500
Subject: Wacky Programming Tales (Was Re: Why use Perl when we've got Python?!)
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net>
Message-ID: <slrn7rdgo2.a5.abigail@alexandra.delanet.com>

Brad Howes (bradh at mediaone.net) wrote on MMCLXXV September MCMXCIII in
<URL:news:wjy7lmxbwh7.fsf_-_ at bradh.ne.mediaone.net>:
&& 
&& Hey! I actually worked on a project where someone did this! Also in their
&& funky include file was
&& 
&&   #define AND &&
&&   #define OR  ||
&& 
&& It was the most bizarre thing I ever saw -- not so much the include file,
&& but the resulting code. As if a Pascal moths had come in, eaten at some
&& code and left little Wirth fragments sprinkled around. Has anyone else
&& encountered such strange programming behavior?

Well, uhm, yeah. I once wrote a C program implementing red-black trees
that almost read like English. Except for parens and the one #include
line, no punctuation chars where needed. It had things like:

      if (I am red and my uncle is black)
      we do   ....

Unfortunally, that code didn't survive an extreme quota regime.



Abigail
-- 
perl -MNet::Dict -we '(Net::Dict -> new (server => "dict.org")
                       -> define ("foldoc", "perl")) [0] -> print'


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From jpeterson at office.colt.net  Mon Aug 23 11:40:44 1999
From: jpeterson at office.colt.net (Jon Peterson)
Date: Mon, 23 Aug 1999 15:40:44 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu> <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net> <slrn7s2lb3.k6b.abigail@alexandra.delanet.com>
Message-ID: <0Ydw3.243$u07.1835@news.colt.net>

In comp.lang.perl.misc Abigail <abigail at delanet.com> wrote:
> xah (xah at weborder.com) wrote on MMCLXXXIII September MCMXCIII in
> <URL:news:37c126a6$0$813 at newsreader.alink.net>:
> "" 
> "" a bit. Their one-language ideology taints our programing world. If you
> "" wonder why so many languages uses weird syntax (e.g. python, dylan, lisp,
> "" haskell, mercury, mathematica ...ad nausea), you have these guys to thank
> "" to.

> Yeah, those math texts from the 16th and 17th century, they are sooooo easy
> to read, because they lack all the symbols. Not to mention the original
> Euclid and those other Greeks, who had no symbols at all!

> "" hope that Larry and Tom's teachings will eventually make such subjects
> "" disappear for good.


> Hahahahahahahahahahhahahahahaahahahahahaha.

Ummmmmmm.... Maybe it's just me but I think Xah was being sarcastic.....

His point being that the restrictions and formalities of, say, Python are
better than the freedoms of, say, Perl, in the same way that the restrictions
and formalities of mathematical notation are better than the freedom of
expression available in plain English.

A glance at Xah's web site indicates that he is not a lover of Perl, and he
is a lover of Mathematics...

 




>   -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
>    http://www.newsfeeds.com       The Largest Usenet Servers in the World!
> ------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From derek at dstc.com  Wed Aug  4 20:52:25 1999
From: derek at dstc.com (Derek Thomson)
Date: Thu, 05 Aug 1999 10:52:25 +1000
Subject: killing a thread
References: <37A8D8DB.4C93D134@cs.utexas.edu>
Message-ID: <37A8E049.504EFE7D@dstc.com>

pascucci wrote:
> 
> Is there a way to kill a thread from the process that generated it?
> I would like to do it both on Windows and Unix (sgi irix) operating
> systems.
> I know that this should be avoided in general but I have
> a computation loop that I wrote in C to optimize speed and for
> performance
> reasons I do not want to put in the inner loop checks for conditions
> that will make the
> thread exit gracefully.
> Valeiro

Then don't put the checks in the inner loop! Or check every n iterations
of the inner loop. Or just let the computation end, and then exit (will
it really waste that many CPU cycles? - you'll have to decide as it
depends on what you're doing).

I couldn't find anything in the thread or threading lib doco to kill a
thread externally, but I presume you looked there already.

In general, I want to warn you away from unilaterally killing a thread
from another thread. It's a bad idea. How do you intend to free any
resources acquired by the thread? Even if you are certain that you've
handled this *today*, tomorrow someone will modify the thread code and
not realize the implications of the fact that the thread could die *at
any time*.

I've never written code that kills a thread directly from another thread
without eventually unpicking it and replacing it with a mechanism that
allows the thread to realize it's time is up, and gracefully shut itself
down.

-- 
__________________________________________________________
Derek Thomson                               derek at dstc.com
DSTC Pty Ltd                               http://dstc.com
University of Qld, 4072                tel +61 7 3365 4310
AUSTRALIA                              fax +61 7 3365 4311
                                     http://dstc.com/Fnorb



From skaller at maxtal.com.au  Fri Aug 20 18:43:30 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Fri, 20 Aug 1999 22:43:30 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au> <7p7h3p$jto$1@newshost.accu.uu.nl>
Message-ID: <37bdceb0.797602261@news.triode.net.au>

On 15 Aug 1999 23:05:29 GMT, m.faassen at vet.uu.nl (Martijn Faassen) wrote:

>John (Max) Skaller <skaller at maxtal.com.au> wrote:

>I'm sure type inference can be done (at least in a limited fashion), but
>it would be nice to have an intermediate dropoff point; a viperi or
>viperc that would work with type annotations. 

	It is always possible to 'do' type inference; this
isn't an issue. The issue is whether I can construct an algorithm
that gives useful (and correct) information, and whether I can
then use that information to optimise performance.

	Type annotations are planned; the syntax will
probably be:

	def f(x:int): pass

because (a) that is fairly standard in other languages, and
(b) it is probably the syntax Guido will adopt. However,
a more transparent scheme is possible (traits a la Swallow; nice idea)

>This is useable already,
>especially if it could be used to create fast Python extension modules.
>(something you aren't aiming for right now, I gather)

	I will look at doing this, but it is unclear if enough
benefits can be obtained: Python is highly polymorphic.

>If the project has usuable results at many stages, you have a bigger chance
>to get the 'open source' effect going (if you're aiming for that), too.

	That is true, but see above. I need to get
much faster performance for interscript, and it isn't
clear the a module compiler can do that. But I will
definitely look at the idea, although the output
would have to be either C/1.5.2 or Java (Jpython).

>Ah, so the reason you don't use Python itself as the interpreter is
>learning, and also as an intermediate dropoff point.

	Not really: C isn't suitable for writing complex
software. I would be mad to use something as low level
as C for programming a compiler/interpreter where 
speed is far less important than the speed of the 
code which is output. Especially while prototyping.

	Secondly, it is easier to extend my own
code than the existing Python code.

>> It is possible the first back end will actually general Ocaml
>> rather than C (since Viper is currently implemented in ocaml).
>
>Hm, why Ocaml? 

	First, it is a powerful language, and is especially
good for implementing language translators.

	Secondly, it can generate assembler output,
so the results are on par with C, sometimes even faster.

	Third, it is based on mathematical foundations.

	it contains many of the best features of
functional, procedural, and object oriented code,
reasonably well integrated, and again _founded_
on mathematics.

	Finally, it is readily available and has a small
footprint.

	To put it in a nutshell, it is easily
the best statically typed programming language
I have seen: and the margin is massive.

>I don't know this language, though I heard some
>interesting things about it. But it begs the question why you're 
>not using Python itself.

	Are you kidding? You really cannot be serious.
Python doesn't have the kinds of constructions needed
for complex software design. Perhaps, after I'm finished
with Viper, it will provide better facilities: one of the aims
is to provide powerful constructions in a unified manner,
but using the nice Python syntax.

>but as you are aiming for a Python compiler
>this can't last. :) And I'd definitely consider at least a C 
>backend for the compiler, because C is so common. 

	I am considering the following backends:

	1) Python 
	2) C
	3) C++
	4) Assembler (via ocaml)
	5) Ocaml

>But Python's C interface is well known and I believe also quite stable;

	No, there was an upgrade from 1.5.1 to 1.5.2,
1.6 may be different, 2.0 will certainly be different.

>I can see where you're coming from, but I'd still seriously consider
>adding at least a CPython interface early on -- the more stages in your
>project that are usable, the higher the chance is your project will
>bear fruit.

	A CPython interface is indeed possible,
but it will be a very _difficult_ problem. The reason
is the the Viper architecture is quite different to
the CPython 1. For a start, there is a built in 
garbage collector.  All the objects are represented
differently. The binding will be _much_ harder
than a completely fresh implementation.

	Note that JPython doesn't support C modules either.

>I can see where full program analysis may help with type inference, 
>of course. But if you go by the intermediate route of manual adding of
>types, it doesn't sound that hard to support some interface with Python.

	True. But that won't work for existing code: I want to
be able to compile _existing_ Python sources.

>Of course you need to restrict the modules that can be called by Viper
>code to the set of modules supported by Viper;

	Yes. So I have to implement all the common library
modules that do not have a Python .py equivalent.
Note that even 'string' has Python source, and,
surprisingly, almost all of  Tkinter is pure python.

>Hm, reading this it seems you're talking about the other thing; supporting
>C Python modules in Viper? I was talking about allowing people to Viper-compile
>their Python modules (as long as they are Viper compliant), and use them
>from Python. But as you say, you can use Pure Python modules in Viper
>as well; if they restrict themselves to the Viper subset. 

	It would be nice to be able to generate compatible
C modules, since that would allow incremental speed up of
existing python programs. 

	However, several competent people have tried this
and the results ar barely better than the bytecode interpreter.

	OTOH experiments in JPython indicate up to 10000X
faster performance after whole program analysis (so I read,
and I believe it).

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From andy.smith at iname.com  Tue Aug 10 16:31:58 1999
From: andy.smith at iname.com (andy.smith at iname.com)
Date: Tue, 10 Aug 1999 20:31:58 GMT
Subject: OCX and PythonWin....
Message-ID: <7oq27u$th0$1@nnrp1.deja.com>

Hi All,

I'm trying to use Pythonwin to host an OCX which provides realtime stock
quotes.

The code I've been using to access the OCX is based on
the 'webbrowser.py'
file in the pythonwin ocx demos directory. The OCX returns quotes to the
application by calling an 'OnDynamicQuoteResponse' method, in response
to a
call to ''SubmitDynamicQuote'. I thought I could catch the event by
mimmicking the way the webbrowser demo catches an 'OnBeforeNavigate2'
event.
Unfortunately though, my 'OnDynamicQuoteResponse' method is not getting
called.

When the OCX is initialised, it requires that the application 'register'
with it. An initial call to the IsRegistered() method returns 0. After
this,
my script calls the Register("MYAPP") method. After this, a further
call to
IsRegistered() returns a one, which signifies that the application has
successfully registered for service. I guess this means that the OCX is
correctly functioning, as I can successfully call its methods, it looks
like
it's just return events that aren't being caught.

Can anyone help?!

Cheers,

Andy.



CODE ##########################################

import win32ui, win32con, win32api, regutil
from pywin.mfc import window, activex
from win32com.client import gencache
import sys

ReutersModule =
gencache.EnsureModule('{B4B75CF0-E748-11D0-9A4F-00A024D831CB}', 0, 1, 0)

if ReutersModule is None:
	raise ImportError, "IE4 does not appear to be installed."

class MyControl(activex.Control, ReutersModule.QOleSvr):
	def OnDynamicQuoteResponse(self, symbol, code, price):
		print "Quote response has been called!!!!"

class ControlFrame(window.MDIChildWnd):
	def Create(self, title, rect = None, parent = None):
		style = win32con.WS_CHILD | win32con.WS_VISIBLE |
win32con.WS_OVERLAPPEDWINDOW
		self._obj_ = win32ui.CreateMDIChild()
		self._obj_.AttachObject(self)
		self._obj_.CreateWindow(None, title, style, rect,
parent)
		rect = self.GetClientRect()
		rect = (0,0,rect[2]-rect[0], rect[3]-rect[1])
		self.ocx = MyControl()
		print "Self.ocx is %s" % self.ocx
		self.ocx.CreateControl("Reuters OCX",
win32con.WS_VISIBLE |
win32con.WS_CHILD, rect, self, 1000)

		print "IsRegistered returns %s" % self.ocx.IsRegistered
()

		self.ocx.Register("MYAPP")
		print "IsRegistered returns %s" % self.ocx.IsRegistered
()

		print "Submitting quote request"
		self.ocx.SubmitDynamicQuote("IBM", "P")
		print "Quote submitted"

if __name__=='__main__':
	f = ControlFrame()
	f.Create("Reuters OCX")


OUTPUT  #################################

PythonWin 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on
win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
Portions Copyright 1994-1999 Mark Hammond (MHammond at skippinet.com.au)
>>> Self.ocx is <win32com.gen_py.QOleSvr OLE Control module.MyControl>
IsRegistered returns 0
IsRegistered returns 1
Submitting quote request
Quote submitted

#################################


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From shecter at darmstadt.gmd.de  Thu Aug 19 07:55:46 1999
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Thu, 19 Aug 1999 13:55:46 +0200
Subject: Kudos to documentors: (popen2)
Message-ID: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de>

Hi,

I needed to do some system programming, and at first tried to use
Perl.  Although I've used Perl for years, this documentation page
forced me to give Python a try for it:

http://www.perl.com/CPAN/doc/manual/html/lib/IPC/Open2.html

(Just notice how much is implicit, and not documented, and with no
links to other needed info.)

I soon found then:

http://www.python.org/doc/current/lib/module-popen2.html

...and in less time than I had wasted with Perl, I was up and running
although I'm relatively new to Python.  The only problem I had at
first (being used to Javadoc) is that the types of the return values
from popen2() weren't explicitly stated, nor were there links to the
docs for them.  But then I just stareted up a Python interpreter, made
a popen2 call, and checked out what I got back.  Not too bad.

Thanks,
Robb



From cgw at fnal.gov  Tue Aug 17 03:48:56 1999
From: cgw at fnal.gov (Charles G Waldman)
Date: Tue, 17 Aug 1999 07:48:56 GMT
Subject: if you want POSIX, import posix, not os
In-Reply-To: <37B8EC03.835F8A51@w3.org>
References: <37B8EC03.835F8A51@w3.org>
Message-ID: <14265.5096.537800.149227@buffalo.fnal.gov>


Dan Connolly writes:
 > I have thought this was broken for years, but I just read
 > it again tonight, and I'm not going to pass over it in silence this
 > time:
 > 
 > "Do not import this module directly. Instead, import the module os,
 > which provides a portable version of this interface."
 > 	-- http://www.python.org/doc/1.5.2p1/lib/module-posix.html
 > 
 > I think this is terrible advice.

I think it's good advice.  If you're not using features like
"getpgrp", but more mundane things like "listdir", then 

import os
files=os.listdir(os.getcwd())

is highly portable (across Mac, Win, and Unix) whereas if you spelled
it

import posix
files=posix.listdir(posix.getcwd())

you'd be introducing a needlessly non-portable construction in your
code.

It's up to you to read the documentation to know what are the
"generic" (always-available) operating system services and which ones
are platform-specific.  The "getpgrp" example you cite is one of the
latter...  if you use it, you  won't be able to get your code to run
on non-Unix platforms.  You have to resort to things like:

import os
if os.name = 'posix':
   pgrp = os.getpgrp()
else:
   #something completely different


 > The os module is *not* portable:
 > 
 > "getpgrp () 
 >     Return the current process group id. Availability: Unix. "
 > 	-- http://www.python.org/doc/1.5.2p1/lib/os-procinfo.html

As the "Availability" comment implies, the os module implements a core 
set of functions, on top of which there are platform-specific
extensions.  If you're not on the right platform, you don't get the
extensions.

 > A portable interface is one where I can expect consistent behaviour
 > of named objects across platforms, no?

Modulo the restrictions of "Availability"

 > The IEEE POSIX spec is widely implemented and understood.
 > When a program expects behaviour as specified in the POSIX
 > specs, it should say
 > 	import posix

If you make heavy use of the posix features this may make sense.  If
all you really need is things like listdir, getcwd, then it's
overkill.

 > It would have made sense for python to define a really portable os
 > module, i.e. the least common denominator of the platforms it
 > runs on. But that's not what we've got.

No reason you can't create an "generic_os" module something like this:

----generic_os.py-----
from os import listdir,getcwd,path,name,remove,....
----------------------

replacing "...." with the list of all functions which are supported on
ALL OS's where Python runs (or all OS's you care about).  A python
program which only imports "generic_os" and not "os" will be highly
portable, and also highly constrained in what it can do.

 > I hope the os module will be deprecated in future releases.

I don't think there is much chance of this happening, it it too widely 
used.





From kjz at kzaragoza.ne.mediaone.net  Fri Aug 20 01:38:10 1999
From: kjz at kzaragoza.ne.mediaone.net (Kris J. Zaragoza)
Date: Fri, 20 Aug 1999 05:38:10 GMT
Subject: Python article on Builder.com
Message-ID: <slrn7rpqbj.atl.kjz@kzaragoza.ne.mediaone.net>

I thought some of you might be interested in Dan Shafer's latest piece
on Builder.Com.  It's a nice bit of PR for Python and might raise its
visibility a little more in the Web app development world.  You can find
the piece at:

http://buzz.builder.com/cgi-bin/WebX?14 at 236.hs95a9MoeFg^0 at .ee7bc67/0

(Now, Dan makes some critical remarks in this article about a few other
languages.  If anyone takes offense at these comments, please direct 
your replies to Dan Shafer directly.  I really don't want to see yet
another language flamewar erupt around here.)

-- 
Kris J. Zaragoza           | "Unfortunately, most people can't out-think a
kzaragoza at mediaone.net     | grapefruit."  --Jon Bodner



From stefan_heinzmann at my-deja.com  Tue Aug 17 08:05:32 1999
From: stefan_heinzmann at my-deja.com (stefan_heinzmann at my-deja.com)
Date: Tue, 17 Aug 1999 12:05:32 GMT
Subject: Fnorb & Tkinter on Win NT, how?
Message-ID: <7pbj67$qro$1@nnrp1.deja.com>

Hi all,

I'm currently trying to use Fnorb 1.01 and Tkinter (Python 1.5.2)
together in a mix-mode client (i.e. client with callbacks) on Win NT
4.0 SP4.

The Fnorb User's Guide says that the two event loops from Fnorb and
Tkinter must be combined to support this. That's why there is a
TkReactor module in the ZIP file, but it doesn't seem to work properly
and the Fnorb User's Guide says it is Unix only.

So has anybody out there made it work on Win platforms? Can't be too
difficult, can it?

BTW, is the Fnorb mailing list dead? I get no reaction and the archives
on the web site are rather stale...

Cheers

Stefan


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From jwbaxter at olympus.net  Wed Aug 25 14:21:50 1999
From: jwbaxter at olympus.net (John W. Baxter)
Date: Wed, 25 Aug 1999 11:21:50 -0700
Subject: xanadu and python
References: <009c01beeed0$4e6babc0$f29b12c2@secret.pythonware.com> <14276.2083.275770.273576@dolphin.mojam.com> <slrn7s82hf.g2c.kaz@ashi.FootPrints.net> <52Uw3.40$2y4.43051@news-dal.corridex.com>
Message-ID: <jwbaxter-2508991121500001@otter.olympus.net>

In article <52Uw3.40$2y4.43051 at news-dal.corridex.com>, "Evan Simpson"
<evan at tokenexchange.com> wrote:

> That 500 Server Error certainly is odd, though.

That typically means what the text says:  there have been more downloads
(probably more bytes, not download actions) than the host allows for that
URL, so it's not going to allow any more until next month or until the
limit is increased.

  --John

-- 
John W. Baxter   Port Ludlow, WA USA  jwb at olympus.net



From thantos at brimstone.mecha  Sat Aug 28 15:06:42 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Sat, 28 Aug 1999 19:06:42 GMT
Subject: Fuzzy string matching?
References: <Pine.LNX.4.10.9908281253390.9521-100000@starship.python.net> <y0jpv07u80k.fsf@vier.idi.ntnu.no>  <y0jlnavu75f.fsf@vier.idi.ntnu.no>
Message-ID: <slrn7sgckn.lbd.thantos@brimstone.mecha>

On 28 Aug 1999 19:50:04 +0200, Magnus L. Hetland <mlh at idt.ntnu.no> wrote:
>
>Yet another version, without the append...

Now, this is how i like to see good code evolve, but one touch ...

>        for j in range(1,n+1):

This is inside the "i" loop; as such, it'll get recomputed every time
through.  Moshe was on the right track to move the computation of the
ranges outside the loop, since this is constat.  :)

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From skip at mojam.com  Tue Aug 31 14:30:43 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 31 Aug 1999 13:30:43 -0500 (CDT)
Subject: Doing POST opperations from Python
In-Reply-To: <37cbeeee.7358361@news.freeserve.net>
References: <37cbeeee.7358361@news.freeserve.net>
Message-ID: <14284.7351.644976.369476@dolphin.mojam.com>

    Phil> I am trying to produce POST requests within python (to drive a web
    Phil> based email client). I can do GET requests, but can understand how
    Phil> to do post. I have read RFC1945 (Informational), but I can find
    Phil> the info there either.  I understand that the data is sent as part
    Phil> of the headers, but how?

Depends which module you're using.  If you use a recent version of urllib,
you can place your parameters in a dictionary, call urllib.urlencode to
encode it and pass the results to urllib.urlopen:

    import urllib
    paramstring = urllib.urlencode({'spam': 1, 'eggs': 2, 'bacon': 0})
    f = urllib.urlopen("http://www.musi-cal.com/cgi-bin/query",
		       paramstring)
    print f.read()

If you're using httplib, you do basically the same thing at a slightly lower 
level:

    import httplib, urllib
    paramstring = urllib.urlencode({'spam': 1, 'eggs': 2, 'bacon': 0})
    h = httplib.HTTP("www.musi-cal.com:80")
    h.putrequest("POST", "/cgi-bin/query")
    h.putheader("Content-length", "%d"%len(paramstring))
    h.endheaders()
    h.send(paramstring)
    reply, msg, hdrs = h.getreply()
    print h.getfile().read()

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From tim_one at email.msn.com  Sat Aug 28 04:05:36 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 28 Aug 1999 04:05:36 -0400
Subject: A Plea For A Return To Sanity!!!
In-Reply-To: <Pine.LNX.4.10.9908271102120.14054-100000@eris.io.com>
Message-ID: <000301bef12c$1bfd6940$ad2d153f@tim>

[Bit Nik]
> I just recently joined this newsgroup/mailinglist and I missed
> out a lot on the previous flame war (lucky me!).

You can relive it in all its dubious glory, at http://www.deja.com/.

> And I certainly don't want to stroke the coals and recreate the
> flame, but i'm very curious /how/ is Python lacking in
> documentation?

It doesn't come in everyone's favorite format.

> This seems to be the major argument against it right?

Doubt it, but it was *an* argument that got an unreasonable amount of
repetition.

> Also, who are these Python biggots people are always talking
> about??

They're more visible on comp.lang.perl.misc, although I'm not sure the Python
community would claim them as their own.  Still, pay your dues to the PSA, and
Guido will give you an alibi if you're ever up on felony charges <wink>.

> ...
> No doubt perl has a lot of cool things going for it too,

It does.

> the perl module collection is unbeatable,

It's certainly large <wink>.

> perl has excellent user support even though there are a few perl
> biggots.

You're certainly more likely to get flamed on c.l.p.m. for asking a "dumb
question", although that's seasonal or something -- it's not always that way,
and most often isn't.

> Most perlers I have met are a preaty logical bunch,

Then you'll find Python a great relief!  Logic doesn't impress anyone here;
paying your PSA dues is all that counts.

> and the perl mongers are on their way to certain world domination.

If *that's* not a sure sign of the beginning of the end, I've lived 300 years
in vain <wink>.

> Another quick question, are there such things as PUGs? (Python User
> Groups). I couldn't find any information on python.org...

SIGs -- Special Interest Groups.  Check the "Community" and "PSA" links -- and
sign up for the PSA while you're there.

python-folk-are-the-best-buddies-psu-dues-can-buy-ly y'rs  - tim





From skip at mojam.com  Mon Aug 30 15:06:12 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 30 Aug 1999 14:06:12 -0500 (CDT)
Subject: Bit fields
In-Reply-To: <7qeim8$3su$1@nnrp1.deja.com>
References: <7qeim8$3su$1@nnrp1.deja.com>
Message-ID: <14282.54432.472029.410470@dolphin.mojam.com>

    Archer> I'm writing some extensions to set up a general SCSI oriented
    Archer> test platform in Python.  Obviously there's a certain amount of
    Archer> C work that has to be done, but I would like to do one of the
    Archer> following:

    Archer> A. access bit fields from a C struct that is returned to Python
    Archer> as a string

This is quite doable.  Python supports the bitwise operators you're familiar
with in C and the struct module lets you pack and unpack C struct info into
and out of Python strings.

    Archer> B. encapsulate the structs as classes with all of the methods
    Archer> implemented in C (this would appear to be possible but the
    Archer> doc is mute).

Also easily doable.  The simplest way I know of is to write an extension
module in C that implements the various methods you are interested in and
then just delegate to them from the actual Python methods, e.g.:

    scsi.py:

	import scsiop		# the extension module containing bit ops

	class SCSI:
	    def __init__(self, ...):
		pass

	    def get_SCSI_id(self, packed):
		"""extract SCSI id from packed string"""
		SCSI_id = scsiop.get_SCSI_id(packed)

In scsiop.c you'd implement interesting functions.  This technique shouldn't
be necessary just to extract bitfields from structs packed into strings, but
may come in handy in other places.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From marc.saric at mpi-dortmund.mpg.de  Tue Aug 31 05:25:53 1999
From: marc.saric at mpi-dortmund.mpg.de (Marc Saric)
Date: Tue, 31 Aug 1999 11:25:53 +0200
Subject: Physical units
References: <FHAq2t.At6@world.std.com>
Message-ID: <37CB9FA1.A11D6A52@mpi-dortmund.mpg.de>

Will Ware schrieb:
> 
> I thought somebody else might find this useful...
> 
> #########################
> '''mks.py - physical units for calculations
> 
Uhm, isn?t that quite similar to the following???

# Physical quantities with units
#
# Written by Konrad Hinsen <hinsen at cnrs-orleans.fr>
# last revision: 1999-7-23
#

# hacked 1998/09/28 GPW: now removes __args__ from local dict after eval
#        1998/09/29 GPW: now supports conversions with offset
#                        (for temperature units)

# $Id: PhysicalQuantities.py,v 1.3.1.7 1999/01/11 21:49:36 gward Exp $

"""Physical quantities with units.

This module provides a data type that represents a physical
quantity together with its unit. It is possible to add and
subtract these quantities if the units are compatible, and
a quantity can be converted to another compatible unit.
Multiplication, subtraction, and raising to integer powers
is allowed without restriction, and the result will have
the correct unit. A quantity can be raised to a non-integer
power only if the result can be represented by integer powers
of the base units.

The values of physical constants are taken from the 1986
recommended values from CODATA. Other conversion factors
(e.g. for British units) come from various sources. I can't
guarantee for the correctness of all entries in the unit
table, so use this at your own risk!
"""

-- 
Bye,

Marc Saric

Max-Planck-Institut f?r molekulare Physiologie
Otto-Hahn-Strasse 11
44227 Dortmund

phone: +49 231 133 2168



From sp00fD at yahoo.com  Tue Aug 24 20:43:01 1999
From: sp00fD at yahoo.com (sp00fD)
Date: Wed, 25 Aug 1999 00:43:01 GMT
Subject: Converting from perl
References: <7pujub$75v$1@nnrp1.deja.com> <m3ogfwrfwo.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <7pve6j$qng$1@nnrp1.deja.com>

In article <m3ogfwrfwo.fsf at atrus.jesus.cam.ac.uk>,
  Michael Hudson <mwh21 at cam.ac.uk> wrote:
> sp00fD <sp00fD at yahoo.com> writes:
>
> > I'm converting to python from perl and just created my first ...

> It might be easier to do this using a dictionary:
>
> arg2hosts = {
>     "AIX":("athena","dev","detroit","la","laredo","miami",
>            "newark","sat","savannah","seattle","titan"),
>     "SUN":("eclipse", "hercules", "scdev","scprod","sol"),
>     "BACKUP":("eclipse", "hercules", "scdev","scprod", "sol",
"athena")
> }
>
> try:
>     self.hosts = arg2hosts[self.args()]
> except KeyError:
>     self.hosts = self.ask_for_hosts()
>

Thanks, I'll do that, the thought crossed my mind, but I couldn't
decide whether to be a memory tightwad or not, you make a good point.

> >                                 os.system("rcp %s %s:%s" % \
> >                                 (_stuff[0], host, _stuff[1]))
> >                 except os.error:
> >                         print "Error during system() function -
oops"
>
> Note that this will not catch errors in executing the command passed
> to os.system - you need to look at the return code for that.

How do I do that?
foo = os.system("..")?

>
> To be honest, I don't see a lot of value in wrapping this up in a
> class. For a straight through shell scripty type job like this I
> generally don't bother. Python is not about ramming methodologies down
> your throat.
>

Honestly, me neither, more than anything I wanted to get a feel for
them, basically I was bound and determined to create a class, but when I
was done, it was obviously stupid.  Unfortunately, most of the stuff I
do is "straight through shell scripty" so, hopefully I can figure out
some decent way to use the OO features of python for that.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From tchrist at mox.perl.com  Wed Aug 18 23:19:56 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 18 Aug 1999 20:19:56 -0700
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <1277462749-32309734@hypernet.com>
Message-ID: <37bb69cc@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    gmcm at hypernet.com writes:
:[posted & mailed]
:Additional point: If x++ were taken as syntactic sugar, it would have 
:the odd affect of modifying a name binding in place. 
:
:If more than syntactic sugar, it would do different things depending 
:on whether x was mutable or immutable.

People are probably only thinking of ++ for numbers, and I don't see that
"in place changes" matter there.  There's no reason that it should make
sense on, say lists.  It's a lot easier to survive without ++ than it
is without +=, however. :-)

--tom
-- 
Even if you do learn to speak correct English, whom are you going to
speak it to?
                --Clarence Darrow



From bowman at montana.com  Sun Aug 29 23:32:09 1999
From: bowman at montana.com (bowman)
Date: Sun, 29 Aug 1999 21:32:09 -0600
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <MPG.1233dfb6145766d998968c@news.mop.no>
Message-ID: <37C9FB39.63A98303@montana.com>

Alexander Staubo wrote:
> 
> A more interesting approach would be to create a Python framework that
> loaded such Parrot scripts and instantiated necessary objects at
> runtime.

I don't know how far I'll take it, but my current thinking is it would
be very convenient to have Tkinter chew through a preexisting .uil file,
and spit out a GUI.  Definitely not platform independent, and so forth,
but, at present, having Tkinter closely resemble C/Motif is one of my
priorities.
   
-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From mlh at idt.ntnu.no  Thu Aug 12 09:43:56 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 12 Aug 1999 15:43:56 +0200
Subject: Macro preprocessor.
References: <slrn7r291c.137.kaz@ashi.FootPrints.net> <37B2B71C.E380A71@appliedbiometrics.com>
Message-ID: <y0joggdf71f.fsf@drue.idi.ntnu.no>

Christian Tismer <tismer at appliedbiometrics.com> writes:

> Kaz Kylheku wrote:
> [description of a cute macro processor skipped]
> 
> > Okay, enough. :) I'll probably get hate mail now about being off topic.
> 
> No hate mail, but some thoughts:
[...]

> There are a few cases where I admit that I generate scriptlets
> on-the-fly using the formatting operator.

Now you've piqued my interest... How do you do this, and in which
cases (for what purposes)? Do you create strings that you execute, or
what? (Sounds like a cool thing to do ;)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From xah at weborder.com  Mon Aug 23 06:48:10 1999
From: xah at weborder.com (xah)
Date: Mon, 23 Aug 1999 03:48:10 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu> <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu>
Message-ID: <37c126a6$0$813@newsreader.alink.net>

Tom Christiansen wrote in message <37b414d7 at cs.colorado.edu>...
>     [courtesy cc of this posting mailed to cited author]
>
>Legibility is purely subjective.
>
>How legible do you find Greek, Arabic, Russian, or Chinese?  And what
>does the character set actually have to do with any inherent complexity
>of the grammar or richness of expressive vocabulary?
>
>Answer: nothing whatsoever.


I'd like to extend what Tom Chrisiansen has said.

Have you ever glanced at a graduate math text? It's probably all Greeks to
you. However, if someone in the know who translated all the symbolism to
plain English, then it becomes legible. It really depends on who's doing the
reading. If you are a math guy, then perhaps symbols is your thing. Most of
us would prefer English. Legibility is purely subjective.

However, the ivy tower class of buttheads insist on symbolism for math,
clinging to the fancy that certain ideas are better expressed in certain
language. They created a whole barrier of entry. You couldn't believe what
kind of perverted use of symbols they invent. I know because I've been there
a bit. Their one-language ideology taints our programing world. If you
wonder why so many languages uses weird syntax (e.g. python, dylan, lisp,
haskell, mercury, mathematica ...ad nausea), you have these guys to thank
to.

I like Perl because it offers freedom and accessibility to imbeciles like
myself. It lets me fuck around, and have my say. I can sit and spin on the
whole established human wisdom. If someone cannot read my code, it is their
shortcoming. As Tom has said in other posts, freedom is a feature. And if
one day I am tired of Larry's Perl, I can fork my own version/vision. (Larry
said it is OK, and in fact encourages forking.)

On a personal note...  Do you know what a computer science degree is these
days? They require you understand shits like lambda calculus, logic, graph
theory, combinatorics, or even algorithms?? All for what? All the Perl
programers I know doesn't understand any of these, and they all are making
big bucks in fortune 500. I'm not ashamed that I dropped out of school. I
hope that Larry and Tom's teachings will eventually make such subjects
disappear for good.

My humble $0.02.

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 "Your nonsense is no better than mine."   --Mathematician of metaland.







From jbauer at rubic.com  Sat Aug 21 23:07:19 1999
From: jbauer at rubic.com (Jeff Bauer)
Date: Sat, 21 Aug 1999 22:07:19 -0500
Subject: ANNOUNCE normalDate 0.7
References: <000301beec48$9d0fa040$cb2d2399@tim>
Message-ID: <37BF6967.81F0A3E0@rubic.com>

Tim Peters wrote:
> OTOH, his does have a Y10K problem, while mine works
> until the year gets so big it won't fit in an unbounded
> Python long, and even then you'll get a butt-saving 
> MemoryError <wink>.
> 
> use-dates.py-if-you're-immortal-ly y'rs  - tim

By then we'll all be speaking Mayan anyway, because the only 
thing left to communicate with will be rocks.

-Jeff



From daniel.dittmar at sap-ag.de  Thu Aug 12 12:00:07 1999
From: daniel.dittmar at sap-ag.de (Daniel Dittmar)
Date: Thu, 12 Aug 1999 16:00:07 GMT
Subject: Extension types as exceptions
Message-ID: <7our22$9fb$1@mailusr.wdf.sap-ag.de>

Should it be possible to use extension types (implemented in C) as
exceptions? 

Currently, this works partly, as you can 'raise' and 'except' them.
It doesn't work in the following code:

try:
    ... code raising exception
except EOFException, err:
    pass

When the exception doesn't match any 'except' clause, it seems to be
reraised through the following code in ceval.c:

		case END_FINALLY:
			v = POP();
			if (PyInt_Check(v)) {
				why = (enum why_code) PyInt_AsLong(v);
				if (why == WHY_RETURN)
					retval = POP();
			}
			else if (PyString_Check(v) || PyClass_Check(v)) {
				w = POP();
				u = POP();
				PyErr_Restore(v, w, u);
				why = WHY_RERAISE;
				break;
			}
			else if (v != Py_None) {
				PyErr_SetString(PyExc_SystemError,
					"'finally' pops bad exception");
				why = WHY_EXCEPTION;
			}
			Py_DECREF(v);
			break;

Only strings and python classes seem to be allowed.

--
Daniel Dittmar
daniel.dittmar at sap-ag.de
SAP AG, Basis Entwicklung Berlin





From sfranke at my-deja.com  Sun Aug 29 12:27:39 1999
From: sfranke at my-deja.com (sfranke at my-deja.com)
Date: Sun, 29 Aug 1999 16:27:39 GMT
Subject: Python for Delphi current?
Message-ID: <7qbn1j$4jp$1@nnrp1.deja.com>

Does the Python for Delphi Package work with
Python 1.5.2 and Delphi 4(5)?

Stefan


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From skip at mojam.com  Tue Aug 24 13:31:46 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 24 Aug 1999 12:31:46 -0500 (CDT)
Subject: Necessity of ``pass''
In-Reply-To: <hhvha5tswc.fsf@forcix.roof.lan>
References: <hhemgu41z7.fsf@forcix.roof.lan>
	<slrn7s408d.gvn.wtanksle@dolphin.openprojects.net>
	<hhvha5tswc.fsf@forcix.roof.lan>
Message-ID: <14274.54686.281949.239343@dolphin.mojam.com>

    forcer> You didn't quote the example where i pointed out the real
    forcer> problem with pass:

    forcer> if foo:
    forcer>     # bar
    forcer>     # baz
    forcer> qux

    forcer> E.g. if you're developing code and just commented out a (not yet
    forcer> worked out branch) to work on a different part of the program.
    forcer> Requireing pass here is unecessary.

A similar problem occurs in try/except statements.  If you want temporarily
disable the try/except so you can see what exception is being raised, you
have to monkey around with the code a bit.  I've gotten used to using "if
0:" and indenting the code another step with C-c > (I use Emacs):

    if foo:
	if 0:
	    bar
	    baz
    qux

The try/except issue is a little messier.  You wind up with something like:

    if 1:
    #try:
	try body
    else:
    #except:
	except body

No language is perfect.  I doubt that eliminating the requirement at least
one statement in a block is going to help much.

There are probably syntactic considerations as well.  If you allow empty
blocks you probably introduce ambiguities in the language that you'd have to 
correct somehow.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From kistler at fnmail.com  Thu Aug 26 03:34:10 1999
From: kistler at fnmail.com (Per Kistler)
Date: Thu, 26 Aug 1999 09:34:10 +0200
Subject: Tkinter, intermediate loop?
Message-ID: <37C4EDF2.E7F2EB82@fnmail.com>

Hi All

If a callback from Tkinters mainloop has been called, can
one therin make an "extra" event loop, to capture all
one needs, and discard the rest? (So to make an application
only respond to a particular toplevel or widget).

Thanks, Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From dwelton at linuxcare.com  Wed Aug  4 18:25:50 1999
From: dwelton at linuxcare.com (David N. Welton)
Date: 04 Aug 1999 15:25:50 -0700
Subject: Get 'ifconfig' information via Python
References: <87yafs8ruu.fsf@linuxcare.com> <37A81400.7474FD89@lemburg.com> <37A82B77.FB3D03E2@callware.com>
Message-ID: <87vhav89ox.fsf@linuxcare.com>

Ivan Van Laningham <ivanlan at callware.com> writes:

> Hi All--
> 
> "M.-A. Lemburg" wrote:
> > 
> > David N. Welton wrote:

> > > res = fcntl.ioctl(s, IN.SIOCSIFHWADDR, ??????)
                              ^^^^^^^^^^^^^

> > Building these structs can be done using the Python struct
> > module... playing around with this can probably crash your system
> > though.

> > In the end, I think you're better off hacking together a
> > new extension module.

This is what I've been working towards... seems easier than futzing
with the 'struct' module stuff.

> Why on earth would reading this information into a struct crash his
> system?  Or am I missing something here?  At worst, writing a
> messed-up struct might hose his ethernet connection, but I really
> doubt that he would be any more likely to crash his system using
> Python than C.

I made a mistake and put SIOC*S*IFHWADDR instead of SIOC*G*IFHWADDR,
which is what I really meant.  

Ciao,
-- 
David N. Welton, Web Engineer, Linuxcare, Inc.
415.354.4878 x241 tel, 415.701.7457 fax
dwelton at linuxcare.com, http://www.linuxcare.com/
Linuxcare. At the center of Linux.



From singlets at my-deja.com  Mon Aug 16 01:31:43 1999
From: singlets at my-deja.com (singlets at my-deja.com)
Date: Mon, 16 Aug 1999 05:31:43 GMT
Subject: How to write to a named pipe?
Message-ID: <7p87nv$fbl$1@nnrp1.deja.com>

Could someone help me with a basic thing?  I have
an external application that reads from a named
pipe and then processes the information from said
pipe.  I'd like to send this info from my python
app.  I can create the named pipe with mkfifo()
just fine.  Reading the docs, I gather that this
only creates the pipe and I must then treat the
pipe as a file. But when I go to access the "file"
(really the pipe), python just hangs.  Really
hangs, i.e., no control-c or control-d stopping.
Here is my simple ditty:

%: python
Python 1.5.1 (#1, Mar 21 1999, 22:49:36)  [GCC
egcs-2.91.66 19990314/Li on linux-i386
Copyright 1991-1995 Stichting Mathematisch
Centrum, Amsterdam
>>> import os
>>> os.mkfifo('/tmp/myfifo')
>>> f=open('/tmp/myfifo','w')    # now we're stuck

I've checked permissions, should be ok.  I've also
tried making the pipe with world rw permissions
from a prompt, then accessing this pipe from
python in the same manner as above.  Same result.
No errors raised, so I assume it's trying to
connect to the pipe.  As you can see, I'm running
python 1.5.1 on a stock RedHat 6 linux Intel box.
Sorry if I'm being an idiot.

Regards,
Steve



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From greg.ewing at compaq.com  Tue Aug 17 18:26:35 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 18 Aug 1999 10:26:35 +1200
Subject: Getting the module name from a method...
References: <7pboln$n74$1@feed.teaser.fr> <37B998BD.1A35F74B@ix.netcom.com>
Message-ID: <37B9E19B.FF90A74A@compaq.com>

"Stephen J. Turner" wrote:
> 
> How about "self.__class__.__module__"?

No, that will give you the module of the
instance's class. This is not necessarily
the same as the class defining the currently
running method, which could be an inherited
class.

Greg



From angela2 at flashmail.com  Sun Aug 15 15:45:28 1999
From: angela2 at flashmail.com (angela2 at flashmail.com)
Date: Sun, 15 Aug 1999 19:45:28 GMT
Subject: Secret Software
Message-ID: <199908160422.NAA26473@dns>

Introducing the HOTTEST selling software of the year.
"NO MORE SECRETS SOFTWARE"  JUST RELEASED! 


The software they want banned.Why? Because these secrets were 
never intended
to reach your eyes !!! 

This is a sophisticated SOFTWARE program DESIGNED that 
automatically
links to thousands of Public Record databases.



*****************************************************************
*****************
SPECIALTY INTERNET ADVERTISING THAT WORKS!
******************************************
ARE YOU TIRED OF GETTING LITTLE OR NO RESPONSE IN YOUR 
ADVERTISING? TRY THE POWER 
OF BULK EMAIL AND CUT YOUR EXPENSES IN HALF. WE CATER TO UPSCALE 
CO'S.ONLY,NO NOVICES.
PRO MAILERS GET IMMEDIATE RESPONSE!  WE CAN TARGET YOUR MARKET. 
TRY THE POWER OF BULK
EMAIL, IT
WORKS OR YOU WOULD'NT SEE THIS! GET STARTED TODAY 708)562-1778..
 
*****************************************************************
*******************  

The NO MORE SECRETS SOFTWARE" is absolutely astounding!  
Here's what you can obtain:

Files that the government has on you.
You know who "The F*I and.......

License plate number!
Get anyone's name and address with just a license plate number!
(Find that girl you met in traffic!)

Driving record! 
Get anyone's driving record

Social security number!
Trace anyone by social security number!

Free Internet Access

Address! 
Get anyone's address with just a name!

Unlisted phone numbers! 
Get anyone's phone number with just a name - even unlisted 
numbers!

Locate! 
Long lost friends, relatives, a past lover who broke your heart!

E-mail! 
Send anonymous e-mail completely untraceable!

Investigate anyone! 
Use the sources that private investigators use (all on the 
Internet)
secretly!

Ex-spouse!
Learn how to get information on an ex-spouse that will help you 
win in court!  (Dig up old skeletons)

Criminal search-background check!
Find out about your daughters boyfriend! 
(or her husband)

Find out! 
If you are being investigated!

Neighbors!
Learn all about your mysterious neighbors!  Find out what they 
have to hide!

People you work with!
Be astonished by what you'll learn about people you work with!

Education verification!
Did he really graduate college?  Find out!

Just insert the floppy disk and Go!

You will be shocked and amazed by the secrets that can be
discovered about absolutely everyone!  Find out the secrets 
they don't want you to know!   About others, about yourself! 

Only $39.95 U.S.  (Postage Paid)

(ORDERS OUTSIDE THE USA, ADD $25.00)

DON'T WAIT TO GET STARTED... here's what to do:


STEP 1 - Compose a new message using the order form text below
STEP 2 - Type or print your order information into the order form 
section
STEP 3 -  Print your completed order, then mail to the address 
below

OPTIONAL, you may choose to place your order on our secure voice 
mail 
system by calling  708-562-1778

Send to:The Resource Marketing Group
 PO Box 7463 Westchester IL. 60154-7463


                        >>> Mail-in Order Form <<<


Name:

Address:

City:

State:

Zip:

Method of payment: 

[   ] Visa [   ] MasterCard  [   ] Check

Credit Card#:
Exp Date:
YOUR SIGNATURE HERE:_______________________________
We cannot process your credit card payment without your signature
(SORRY, NO MAC VERSION AVAILABLE AT THIS TIME.)
NOTE: THIS PROGRAM WILL WORK ON WINDOWS 3.11 AND OLDER

DISCLAIMER:  The distributor of this powerful software resource 
will be not be held responsible for how the purchaser chooses to 
use its resources.

*****************************************************************
*********************************
To be removed from this listserver,
Please include the word unsubscribe in the subject header!!! 
angela2 at flashmail.com

 
 
 
 
 
 
 
 
 
 
 



From philh at vision25.demon.co.uk  Sun Aug  8 17:27:06 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 08 Aug 99 21:27:06 GMT
Subject: Problem with modules refering to each other
References: <000701bee1c2$25b868a0$722d2399@tim>
Message-ID: <934147626snz@vision25.demon.co.uk>

In article <000701bee1c2$25b868a0$722d2399 at tim>
           tim_one at email.msn.com "Tim Peters" writes:
> [Phil Hunt, mutual imports]
> 
> Have you looked at the FAQ?  4.37, "How can I have modules that mutually
> import each other?".

Unfortunately the FAQ's been removed from the python distro that comes 
with SuSE6.0.

I'll have a look at work tomorrow (where I have high-speed web access). 

-- 
Phil Hunt....philh at vision25.demon.co.uk




From rra at stanford.edu  Mon Aug 16 06:42:43 1999
From: rra at stanford.edu (Russ Allbery)
Date: 16 Aug 1999 03:42:43 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com> <37B7E1AC.FF4FF2CF@strs.co.uk>
Message-ID: <ylso5kqa58.fsf@windlord.stanford.edu>

In comp.lang.perl.misc, Ian Clarke <I.Clarke at strs.co.uk> writes:

> I agree completely.  As the person who started this whole thing
> (sheepish grin) the regulars, Tom in particular, has been rude, and
> resorted to personal insults on a number of occasions, even while I
> remained calm and polite.  He repeatedly accuses me of being a troll and
> inciting flames (despite the fact that he was the only one flaming!).

However rude Tom has been, apply a little bit of logic to what you're
doing.  Crossposting between Python and Perl newsgroups is inciting
flames.  Starting threads with the topic "Why use Python when we've got
Perl?" or "Why use Perl when we've got Python?" or the like is inciting
flames.  Starting threads about Python in a Perl newsgroup or Perl in a
Python newsgroup is inciting flames.

Sometimes it's worth putting up with the flames in order to make a point
or to get some useful information, and I think there's been some useful
and interesting information in this thread.  But please take some personal
responsibility for your actions.

comp.lang.perl.misc is primarily for talking about Perl, not trying to
defend Perl to a Python advocate.  I'm sure the converse is true of
comp.lang.python.  Someone who starts a language advocacy flamewar and
then plays innocent about it isn't exactly holding the moral high ground.

-- 
#!/usr/bin/perl -- Russ Allbery, Just Another Perl Hacker
$^=q;@!>~|{>krw>yn{u<$$<[~||<Juukn{=,<S~|}<Jwx}qn{<Yn{u<Qjltn{ > 0gFzD gD,
 00Fz, 0,,( 0hF 0g)F/=, 0> "L$/GEIFewe{,$/ 0C$~> "@=,m,|,(e 0.), 01,pnn,y{
rw} >;,$0=q,$,,($_=$^)=~y,$/ C-~><@=\n\r,-~$:-u/ #y,d,s,(\$.),$1,gee,print



From robind at earthling.net  Mon Aug  2 12:46:23 1999
From: robind at earthling.net (Robin Dunn)
Date: Mon, 2 Aug 1999 09:46:23 -0700
Subject: frozen apps on AIX
Message-ID: <006801bedd06$977593c0$1a25d2d1@jenkondev.com>

I have an app I am trying to freeze on AIX.  It runs fine unfrozen, but as
soon as the frozen version tries to import an extension module in a .so it
core dumps.
It seems to be in the initial startup parts of the module's init function,
but nowhere that makes sense, somewhere between another function's return
and control coming back to the init function.

--
Robin Dunn
Software Craftsman
robin at AllDunn.com
http://AllDunn.com/robin/
http://AllDunn.com/wxPython/  Check it out!





From gfeeney at hawaii.edu  Mon Aug 23 05:26:18 1999
From: gfeeney at hawaii.edu (Griffith Feeney)
Date: Sun, 22 Aug 1999 23:26:18 -1000
Subject: can python/win95 run interactive dos program?
Message-ID: <3.0.5.32.19990822232618.0082b210@pop-server.hawaii.edu>

I'd like to call a dos program that prompts for two simple text string
inputs from a python program running under win95, using some expect-like
facility. Possible?




From skaller at maxtal.com.au  Thu Aug  5 17:39:07 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Thu, 05 Aug 1999 21:39:07 GMT
Subject: International characters and Python's string functions on Linux
References: <MPG.1213b1f2ede568ee989681@news.mop.no>
Message-ID: <37aa0196.1944357548@news.triode.net.au>

On Thu, 5 Aug 1999 15:37:13 +0200, earlybird at mop.no (Alexander Staubo) wrote:

>I'm running Python on Slackware Linux, and it does not seem to provide a 
>full international character set in the string.uppercase and 
>string.lowercase constants: only A-Z, which isn't enough.
>
>Having studied the string module code a bit, I realize that Python is at 
>the mercy of the C API and its string/locale functions, so I'm thinking I 
>must probably configure Linux for a specific locale. Right? If so, how?

	Wrong. Python doesn't use C locale functions.
Also, it doesn't support ISO-10646 directly. If you wish to support
any internationalisation, the recommended approach is to conform
to ISO-10646, which is an International Standard.

	You can write your own code now, and you can borrow
mine: see

	http://www.triode.net.au/~skaller/unicode/index.html

for documented sources. 

	In Python today, you can support
internationalisation using the UTF-8 encoding, as my 
literate programming tool interscript does. UTF-8 is,
in my opinion, the best option in Python today, since it
is ASCII compatible, and will work with 8 bit strings as 
Python today already has.

	The LAST thing you should do is give any
internal support to ISO-8859-x encodings such as Latin-1
etc, which are NOT UTF-8 compatible, nor are
they compatible with each other. In other words
do NOT use any single (or double) byte international 
character set encodings. Use UTF-8, or use arrays
of integers representing the 31 bit UCS-4 
(direct) encoding of ISO-10646.

	By the way, UTF-8 is the official
Linux encoding. Try 

	man utf-8

and replace the incorrect word 'unicode' with
ISO-10646 in the man page: UTF-8 provides
a full encoding of all 2^31 codepoints.
[Unicode is the first 2^16 of them]

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From gmcm at hypernet.com  Sun Aug  8 11:54:35 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 8 Aug 1999 10:54:35 -0500
Subject: ANN: Lightweight C++ interface classes
Message-ID: <1277998619-111245@hypernet.com>

SCXX is a very lightweight set of C++ classes that wrap the most 
commonly used Python objects.

It is inspired by Paul Dubois's CXX (part of the LLNL download), but 
does not use templates or STL, (it does use exceptions) and only 
covers a subset of CXX's capabilities. (And this is why I wrote SCXX 
- templates / STL don't work or bloat on many compilers; and I wanted 
something very lightweight.)

Like CXX, the primary advantage is that you can write C++ code that 
looks a whole lot like Python. Reference counts are taken care of; 
wrapping an existing object forces a typecheck; you can use square 
brackets to access sequences and dicts, just like you'd expect...

SCXX uses the highest possible level of the Python / C API at all 
times. While this is slow (refcounting has not been optimized), it is 
very small, and most of it will be inlined on most compilers.

SCXX lives on my starship pages
(starship.python.net/crew/gmcm/scxx.html). It is used by the new
MetaKit wrap that Jean-Claude just announced. (The backlink to
MkWrap is broken, I'll fix this when my ISP's DNS servers come
back).

SCXX has been tested on Windows (MSVC 5) and Linux (egcs) by me, and
on MSVC 6 and newer releases of egcs by Jean-Claude.

[RH 5.2's distribution of egcs (egcs2.90.29 - egcs-1.0.3) is broken 
when it comes to exceptions (catch doesn't). This is fixable by 
downloading the sources and rebuilding.]

The license is "do as thou wilt", but please maintain the copyright 
notices; and no warranty, of course.

- Gordon



From tiddlerdeja at my-deja.com  Sun Aug 22 06:27:40 1999
From: tiddlerdeja at my-deja.com (tiddlerdeja at my-deja.com)
Date: Sun, 22 Aug 1999 10:27:40 GMT
Subject: passwd/MD5 help?
Message-ID: <7pojao$2pv$1@nnrp1.deja.com>

Can you help?
	 I have a 3rd party program that takes a program, encrypts it and
stores it in an Oracle database. I'd like to know what mechanism the
program uses to encrypt the password. This isn't to crack the passwords
but to mirror the mechanism it uses so that I can write a similar
program to insert parwords into the database. I prehaps thought it was
using MD5 then uuencoding the result. This is not the case as the
password in stored as 20 ascii chars.

If you have any pointers to a solution I'd really appreciate it.

Thanks,

Kieran breen


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From pduffin at mailserver.hursley.ibm.com  Fri Aug  6 11:33:16 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 06 Aug 1999 16:33:16 +0100
Subject: Linux why not fully dynamic?
References: <1278177653-25589496@hypernet.com> <37AAF3D5.141B55EB@bioreason.com>
Message-ID: <37AB003C.28D9@mailserver.hursley.ibm.com>

Mitch Chapman wrote:
> 
> Gordon McMillan wrote:
> >
> > Robin Becker asks:
> >
> > > Why doesn't ./configure set up for a libpython.so.1.5.2 or whatever?
> >
> > Don't know. Maybe because this kind of scheme doesn't work on
> > some other Unixen?
> 
> This question (or something similar to it) is answered in the Python
> FAQ:
>         http://www.python.org/doc/FAQ.html#3.30
> 
> -----------------------------------------------------------------------
> 3.30. Why is the Python interpreter not built as a shared
> library?
> 
> (This is a Unix question; on Mac and Windows, it is a shared library.)
> 
> It's just a nightmare to get this to work on all different
> platforms. Shared library portability is a pain. And yes, I know about
> GNU libtool -- but it requires me to use its conventions for filenames
> etc, and it would require a complete and utter rewrite of all the
> makefile and config tools I'm currently using.
> 
> In practice, few applications embed Python -- it's much more common to
> have Python extensions, which already are shared libraries. Also,
> serious embedders often want total control over which Python version
> and configuration they use so they wouldn't want to use a standard
> shared library anyway. So while the motivation of saving space when
> lots of apps embed Python is nice in theory, I doubt that it will save
> much in practice.  (Hence the low priority I give to making a shared
> library.)
> -----------------------------------------------------------------------

AFAIK this is a quote from GvR which was given a long time ago, I would
be interested to know if he has changed his mind on this.

Whenever anyone asks this question they "eventually" get pointed to this
quote. I think that there have been enough people asking about this to
invalidate the primary reason for this being given such a low profile.

Considering the fact that Python already has support for dynamic
extensions and that it uses backlinking which does not care whether
the symbols are in a library or an executable it is not as difficult
as Guido thinks.

Unfortunately because of this statement people are reluctant to do
the necessary work to support this because they do not feel that it
will be added into the code. Instead they workaround it or do enough
for themselves and so the problem never gets fixed.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From skaller at maxtal.com.au  Mon Aug 23 22:29:31 1999
From: skaller at maxtal.com.au (John Skaller)
Date: Tue, 24 Aug 1999 12:29:31 +1000
Subject: Extending Python: exception handling
In-Reply-To: <14271.18469.757616.424717@dolphin.mojam.com>
References: <37bdcb30.796706964@news.triode.net.au>
 <37bdcb30.796706964@news.triode.net.au>
Message-ID: <3.0.6.32.19990824122931.0099e860@mail.triode.net.au>

At 19:50 21/08/99 -0500, you wrote:
>
>    John> I would like to know how YOU are using exception handling OTHER
>    John> than for error management.
>
>It makes a reasonable substitute for the lack of labelled break statements
>on certain occasions.
>
>    class BreakLabel: pass
>    try:
>	for i in ...:
>	    for j in ...:
>		exit_cond = do_some_work()
>		if exit_cond: raise BreakLabel
>    except BreakLabel:
>	pass

Two questions:

	1) would you always use a user defined exception?
	2) What if you had labelled breaks?

Reason: suppose I implement labelled statements in Viper,
and suppose I assume that system exceptions (Attribute
errors, etc) are program faults -- would all your code work?

The above code would work, because it uses a user defined
exception. But consider:

	try:
		y = x.sqrt()
		print "Got class instance sqrt"
	except AttributeError:
		y = sqrt(x)
		print "Got sqrt of float"

This code might 'fail' under Viper, since 'sqrt' is a method
of floats. Even worse, in some cases a _compile time error_
might be detected by type inference. To illustrate:

	try:
		y = sqrt(x)
	except:
		y = x.sqrt()

Here, the programmer is assuming that if the builtin sqrt
function defined for floats fails, the object might be a
class emulating numbers, and try to call a sqrt method.
But the Viper compiler might _know_ that x is a class instance
and that the builtin function 'sqrt' only applies to floats,
and report an error. 

	Even worse, it might _deduce_ that x must be a float,
and then get confused seeing x.sqrt(), which might require
that x is an instance (depending on whether sqrt is a method
of floats or not).

	The basic problem is that in the presence of
dynamic typing and exception handling, type inference
is hard, especially because handlers for exceptions can
be lexically isolated from the point the exception is raised:
with a _complete_ flow analysis, the type inference system
wouldn't work in all cases. [and maybe not even then]


-------------------------------------------------------
John Skaller    email: skaller at maxtal.com.au
		http://www.maxtal.com.au/~skaller
		phone: 61-2-96600850
		snail: 10/1 Toxteth Rd, Glebe NSW 2037, Australia





From skip at mojam.com  Sun Aug 29 22:44:32 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sun, 29 Aug 1999 21:44:32 -0500 (CDT)
Subject: Returning none
In-Reply-To: <7qbq25$kj6$1@newshost.accu.uu.nl>
References: <slrn7sgesi.lbd.thantos@brimstone.mecha>
	<99082815301626.02087@quadra.teleo.net>
	<7qbq25$kj6$1@newshost.accu.uu.nl>
Message-ID: <14281.61423.566699.84062@dolphin.mojam.com>

    Martijn> Patrick Phalen <python-list at teleo.net> wrote:
    >> On Sat, 28 Aug 1999, Alexander Williams wrote:

    >>> There are no procedures in Python.  There are only function calls (and
    >>> Zuul).

    >> Zuul?

    >> Must be Dutch ...

    Martijn> It ain't Dutch. :)

    Martijn> Probably-some-PSU-codeword-ly yours,

Perhaps the lord of the underworld from Ghostbusters?

S



From bowman at montana.com  Sat Aug 21 13:03:47 1999
From: bowman at montana.com (bowman)
Date: Sat, 21 Aug 1999 11:03:47 -0600
Subject: Man pages would be cool.
References: <37BC6986.A66B044E@starmedia.net>
					<199908191634.KAA01333@jhereg.perl.com>
					<14268.21198.934117.241668@anthem.cnri.reston.va.us>
					<14268.23007.239507.166209@weyr.cnri.reston.va.us>
			<y0j7lmq44wd.fsf@vier.idi.ntnu.no>
			<37BDD1B9.99588CB@prescod.net> <37BE24E0.79EA064E@montana.com> <37BE9EAB.CB7B1380@prescod.net>
Message-ID: <37BEDBF3.BCA64E58@montana.com>

Paul Prescod wrote:
> 
> I'm talking more from a marketing perspective. Wouldn't it be great if
> new Linux users just walking through the menus would "stumble" upon
> IDLE? 

And try to use it like the Notepad type editor it resembles at first
glance? I understand what you are saying, but I don't think chance
encounters are very productive. At least IDLE provides a readily
apparent escape hatch, unlike stumbling into emacs in the pre-menu days. 

-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From abigail at delanet.com  Sun Aug 15 06:25:34 1999
From: abigail at delanet.com (Abigail)
Date: 15 Aug 1999 05:25:34 -0500
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cf10@cs.colorado.edu>
Message-ID: <slrn7rd5et.a5.abigail@alexandra.delanet.com>

John Stevens (jstevens at bamboo.verinet.com) wrote on MMCLXXIV September
MCMXCIII in <URL:news:37b4cf10 at cs.colorado.edu>:
__ In comp.lang.perl.misc, you wrote:

Who is "you"?

__ 
__ >A CGI monstrosity from a 
__ >script kiddie isn't, but this is irrelevant.
__ 
__ Actually, they are *VERY* relevant!  Those CGI monstrosities are
__ Perl, they go into the statistics, and somebody somewhere has to
__ maintain them, and somebody has to pay to get them maintained.


So, the best language would be one noone uses? So, it can't be misused
and held against you?

Brilliant idea. Very useful as well.


Abigail
-- 
               split // => '"';
${"@_"} = "/"; split // => eval join "+" => 1 .. 7;
*{"@_"} = sub {foreach (sort keys %_)  {print "$_ $_{$_} "}};
%{"@_"} = %_ = (Just => another => Perl => Hacker); &{%{%_}};


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From tre17 at pc142.cosc.canterbury.ac.nz  Thu Aug 19 20:46:45 1999
From: tre17 at pc142.cosc.canterbury.ac.nz (Timothy R Evans)
Date: 20 Aug 1999 12:46:45 +1200
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <c75yafced1p.fsf@pc142.cosc.canterbury.ac.nz> <37bb6923@cs.colorado.edu>
Message-ID: <c75g11f1doq.fsf@pc142.cosc.canterbury.ac.nz>

Tom Christiansen <tchrist at mox.perl.com> writes:

>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.python, 
>     Timothy R Evans <tre17 at pc142.cosc.canterbury.ac.nz> writes:
[snip]
>     def foo():
> 	x = x + 2
> 	return x
> 
> That's not legal even now.  You need the global.
> 
[snip]

Oops, you're right of course, you can access globals but you can't
override a global definition if you have already used the global
value.  Quite strange behaviour really.

x = 5
def foo():
    print x # works, prints 5
def bar():
    x = 6 # works, but doesn't change the global
def wibble():
    print x
    x = 6 # this fails with NameError

Seems like a good behaviour as it stops you doing dumb things.

--
Tim Evans




From mstenber at cc.Helsinki.FI  Thu Aug  5 00:12:57 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 05 Aug 1999 07:12:57 +0300
Subject: Any clues to source of this delay?
References: <al8vhaxtg1y.fsf@sirppi.helsinki.fi> <14247.22099.356207.503879@dolphin.mojam.com> <al8d7x4qkoc.fsf@sirppi.helsinki.fi> <0E16861EE7BCD111BE9400805FE6841F0AB89844@c1s5x001.cor.srvfarm.origin-it.com>
Message-ID: <al8so5yoofq.fsf@sirppi.helsinki.fi>

"Robin Boerdijk" <robin.boerdijk at nl.origin-it.com> writes:
> > Oh, sorry, wasn't being explicit; it isn't slow start, as the ceiling is
> on
> > _single_ connection, however long it stays up. By opening multiple
> > connections, the server's total handling capacity goes up but single
> > connections' speed stays constant.
> At the risk of being completely wrong: your problem might be caused by the
> fact that TCP has a sliding window of 64Kb. This means that there can never
> be more than 64Kb of outstanding, unacknowledged data on a single TCP
> connection. When TCP reaches that limit, it will stop sending data until an
> acknowledgement from the receiver makes the window 'slide' to the right. The
> larger the size of the packets you sent, the sooner the windows fills up and
> the less rountrips per second you will get.

Data transferred was minimal.. _apparently_ it was mainly caused by delayed
ack, although, I doubt I'll ever know for sure (too lazy to reread RFC's;
delayed ack would fit the picture though, or possibly slow start without
window size growing).

> Hoping this guess is not totally off,
> 
> Robin.

-Markus

-- 

	Only in the silence voice, Only in the darkness light, 
	Only in dying life; The hawk's bright flight on the empty sky
		-- Ursula Le Guin / (the) Earthsea Quartet



From fw at s.netic.de  Fri Aug 20 06:29:43 1999
From: fw at s.netic.de (Florian Weimer)
Date: 20 Aug 1999 12:29:43 +0200
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7rpksg.e6c.wtanksle@dolphin.openprojects.net>
Message-ID: <87k8qqago8.fsf@deneb.cygnus.qad.org>

wtanksle at dolphin.openprojects.net (William Tanksley) writes:

> 
> >                                Note that Python doesn't support hash
> >                                slices, which in Perl are simple:
> >                                    @hash{k1,k2,k3} = (v1, v2, v3)
> 
> That's a good point, although I'm not sure what Python can do about it --
> our slice semantics talk about gaps, and hashes don't have gaps.  I
> suspect we'd have to call it something other than slicing (this seems
> pretty reasonable).

You can use `hash.update({k1: v1, k2: v2, k3: v3})', which has the
additional benefit of putting keys and values closely together.



From banderson at boi.hp.com  Fri Aug 27 15:17:10 1999
From: banderson at boi.hp.com (Bill Anderson)
Date: Fri, 27 Aug 1999 13:17:10 -0600
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net> <slrn7s2lb3.k6b.abigail@alexandra.delanet.com> <7putsq$3ql$1@lycaeus.calstatela.edu> <MPG.122fcf84a811c6099896e8@news.melbpc.org.au>
Message-ID: <37C6E436.CC3338D9@boi.hp.com>

Factory wrote:
> 
> In article <7putsq$3ql$1 at lycaeus.calstatela.edu>, ramune at bigfoot.com
> says...
> > In article <slrn7s2lb3.k6b.abigail at alexandra.delanet.com>,
> > Abigail <abigail at delanet.com> wrote:
> > [snip amusing rant]
> > >Yeah, those math texts from the 16th and 17th century, they are sooooo easy
> > >to read, because they lack all the symbols. Not to mention the original
> > >Euclid and those other Greeks, who had no symbols at all!
> >
> > <GRIN>
> > Oh, but they're written with only symbols!  All those confusing greek symbols
> > that mathematicians use are the basic building blocks for the language!  Why
> > didn't they just use English?
> > </GRIN>
> 
>   Hmm that might be a bit off, the anchient Greeks had no concept of
> variables, thus you would never see f(x)= 2y in any old greek
> mathematics. IIRC variables were discovered(?) by Al-Jabar sometime
> after.

Correct by my memory (whence came Algebra).

>   Thus the Greeks certainly did not use symbols, and it held back
> mathematics. Not that this applies to python vs. perl.

Which of course has no real relation to c.l.python ;-)

-- 
Bill Anderson 					Linux/Unix Administrator, Security Analyst
ESBU (ARC)					banderson at boi.hp.com
My opinions are just that; _my_ opinions.



From wtanksle at dolphin.openprojects.net  Sun Aug 15 19:54:00 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Sun, 15 Aug 1999 23:54:00 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au> <7p7h3p$jto$1@newshost.accu.uu.nl>
Message-ID: <slrn7rekon.r7j.wtanksle@dolphin.openprojects.net>

On 15 Aug 1999 23:05:29 GMT, Martijn Faassen wrote:
>John (Max) Skaller <skaller at maxtal.com.au> wrote:

>[type inference, me saying it might be hard, John saying it is doable]

Type inference and native code compilation might be simultaneously
possible if you write Viper (or Swallow) in Pop-11, which in spite of
being a general-purpose language has manyamenities for language
programmers.  Common Lisp, ML, and Prolog are already written in it, and
all compile to native code (and work interactively).

Best of all, Pop-11 is finally free.

>> It is possible the first back end will actually general Ocaml
>> rather than C (since Viper is currently implemented in ocaml).

>Hm, why Ocaml? I don't know this language, though I heard some
>interesting things about it. But it begs the question why you're 
>not using Python itself. One reason may be because Python is too
>slow to build an interpreter, but as you are aiming for a Python compiler
>this can't last. :) And I'd definitely consider at least a C 
>backend for the compiler, because C is so common. Though Ocaml may be
>okay that way too, I don't know.

ML does type inferance.  He's probably using that just so  he gets the
inference free.

>Martijn

-- 
-William "Billy" Tanksley



From aahz at netcom.com  Mon Aug  9 12:04:00 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 9 Aug 1999 16:04:00 GMT
Subject: Licenses...
References: <37A6A2DB.9B11FB01@strs.co.uk> <37AE9AC3.34007C4E@lemburg.com> <37AE9B42.4DF40758@lemburg.com> <SpY1dCA7Ctr3EwsH@jessikat.demon.co.uk>
Message-ID: <7omu5g$23h@dfw-ixnews9.ix.netcom.com>

In article <SpY1dCA7Ctr3EwsH at jessikat.demon.co.uk>,
Robin Becker  <robin at jessikat.demon.co.uk> wrote:
>In article <37AE9B42.4DF40758 at lemburg.com>, M.-A. Lemburg
><mal at lemburg.com> writes
>...
>>If you link against a GPLed product (even if it's done dynamically!),
>>the whole construction (your product + the GPLed one) automatically
>>fall under the GPL. This means that you can still sell and distribute
>>your product, but only if you include the full source code of
>>the GPLed product
>>*plus* your product.
>
>Oh dear; I wrote this extension called dll for Tcl that allows tcl code
>to link to arbitrary dll's. The source code for the extension is
>available. Must I distribute source for all known GPL'd products which
>can be linked to tcl using this extension. As it happens this thing is
>mostly used by win32ers, but it has run under Linux.

IANAL, but I believe that it's only distribution of a package that
triggers these clauses.  Thus, as long as you only distribute your own
code, you are only responsible for your license plus the licenses of
whatever packages you *need* in order for your package to be used; you
are not responsible for downstream uses of your package.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From mwh21 at cam.ac.uk  Wed Aug 11 13:23:06 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 11 Aug 1999 18:23:06 +0100
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au>
Message-ID: <m3vham44g5.fsf@atrus.jesus.cam.ac.uk>

skaller at maxtal.com.au (John (Max) Skaller) writes:

> This is a request for comments on Viper, Yet Another implementation
> of Python. I'm building Viperi, the interpreter,  and I'm interested in
> comments on three issues:
> 
> 	1) compatability with Python
> 	2) Extensions
> 	3) Optimisation
> 
> Viperi is being built as a prelude to implementation of an optimising
> compiler, Viperc. Unlike previous attempts at a compiler, Viperc will do
> type inference, inlining, and special case analysis,
> in an attempt to optimise generated code.

I think JPython does some global analysis; not heard anything on that
front for a while though.

I'm interested (do you have any code yet?) but I have a few questions:

> The interpreter aims to be compatible with 'pure python':
> python not using external C modules, and not fiddling
> implementation details too much. Some of the fiddles
> will work, and some will not. 

I presume fiddling with sys.exc_traceback.f_back.f_locals is out?

> Compatibility:
> 	1) exceptions: both string and class exceptions are supported,
> 	the class exceptions are client defined in exceptions.py
> 
> 	2) rebinding module variables after importing is not permitted

Does this mean the following is illegal:

import foo
foo.bar = 1

But can foo then have a method set_bar? I guess that would be easier
to cope with for the analyser.

> 	3) None is a keyword.
> 	4) range is a keyword

What about other builtins? Can't your static analysis find rebindings
of builtin names (including range) rather easily? - particularly as
the don't happen very often.

> 	5) What restrictions can you live with?

One thiung I occasionally find very handy is the ability to assign
methods to classes (or more usually within classes).

> Extensions:
> 	1) full slicing semantics

Cool.

> 	2) optional values in dictionaries (Defaults to None)

Now this one I don't get; do you mean dict[] has dict.get like
semantics?

> 	3) 'in' applies to dictionaries 'as if a sequence of keys were used'

Seems unecessary to me. Maybe that's just me.

> 	4) lexically scoped functions (including lambdas)

Cool.

> 	5) optionally typed function parameters

Out of curiosity, what syntax do use for this? I presume if a
parameter is typed as `Foo' then a value of type `Bar' where Bar
derives from Foo is allowable.

> 	6) what do you want?

Can you derive from basic types? That might make some stuff harder.

[snip]
> 
> 	It is difficult to optimise individual
> modules. It is a different story to optimise
> a whole program, where _every_ call
> to a function can be traced. Of course,
> this may involve restricting what 'exec' can do,

Like banning it entirely, I suspect.

> and it may involve other restrictions
> (such as not  changing any module bindings
> after loading).

Is that really that much of an issue? I'd have thought that they'd be
fairly easy to spot. Mind you, the whole thing sounds sooo hard, that
anythong that makes it easier is good.

Good luck! (I think you may need it)

Michael



From cbbrowne at news.hex.net  Fri Aug 13 21:24:14 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Sat, 14 Aug 1999 01:24:14 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk> <luemh851ym.fsf@stest.ch> <YeMs3.6620$gO1.175350@news2.giganews.com> <m3u2q42ole.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <2z3t3.7359$gO1.213382@news2.giganews.com>

On 13 Aug 1999 07:15:25 +0100, Michael Hudson <mwh21 at cam.ac.uk> wrote:
>cbbrowne at news.hex.net (Christopher Browne) writes:
>> On 12 Aug 1999 19:43:45 +0200, Christian Egli <christian.egli at alcatel.ch> wrote:
>> >>>>>> "Phil" == Phil Hunt <philh at vision25.demon.co.uk> writes:
>> >>>>>> "Paul" == Paul Prescod <paul at prescod.net> writes:
>> >Paul> Glade is a GUI application that allows you to build a window
>> >Paul> visually.
>> > 
>> >Phil> Okay, but I saw a reference somewhere to Glade having a
>> >Phil> declarative file format that could be converted into code to
>> >Phil> build the GUI. I was mostly talking about the declarative file
>> >Phil> format.
>> >
>> >Maybe you meant the storage format of Glade which is XML which in
>> >turn can be used with Libglade to build the GUI dynamically.
>> 
>> Sounds Really A Lot like CORBA DII...
>
>Not to me. Could you elaborate for the slow of thought?

DII == Dynamic Invocation Interface.

Basic idea is that it allows a CORBA client to dynamically discover
new objects and their interfaces, and connect to them.

Same idea as having a LibGlade client dynamically discover new GTK
widgets and their interfaces, and then connecting to them.

Upside: It's pretty cool, and potentially makes programs more
flexible.

Downside: You have to write the client so that it knows how to
discover and use these new objects.  Suffice it to say that This Can
Get Complex.

>> >Paul> Glade does that too, as long as your "favorite language" is
>> >Paul> C. :)
>> >
>> >Not only. See below
>> ...
>> >AFAIK Libglade does exactly that.
>> 
>> Question: How do you tie events to widgets?
>> 
>> Is it, maybe, perchance, via pointers to C functions???
>
>widget.connect("<signal>",<callback>).
>
>I mean, yes, gtk+ ties events to widgets via C function pointers, but
>the Python wrappers are sufficiently sophisticated that you can use a
>Python function. 

Glad to hear it.

>pygtk rocks. I'm not sure it's as perfectly designed as could be, but
>it's pretty damn cool.
-- 
Non-determinism means never having to say you're wrong.
cbbrowne at ntlug.org- <http://www.hex.net/~cbbrowne/lsf.html>



From guy_oliver at yahoo.com  Sun Aug 29 11:17:53 1999
From: guy_oliver at yahoo.com (guy_oliver at yahoo.com)
Date: Sun, 29 Aug 1999 15:17:53 GMT
Subject: Why do we call python a scripting language?
References: <Pine.LNX.4.04.9908290357540.16114-100000@gnr.u2me3.com>
Message-ID: <37c94c0c.1471116@news.netset.com>

On Sun, 29 Aug 1999 04:00:14 -0400 (EDT), Stephen
<stephen at webadmins.com> wrote:

><snipped for brevity>
>'not a real language capable of real programs.'??
>
> Are you on crack? Or do you have a different definition of the 
> word 'real'? Perl and python or both 'real' languages. I can
> write just about anything from sockets to log parsing in 
> them with the greatest of ease.

 You missed the point.  I said 'others' keep suggesting to me that
python is not appropriate for some job or the other since they hear
its a scripting language and equate scripting with simple programs.
Its mostly because their experience with scripts is limited to simple
things.  I recognize that most scripting languages can do a lot of
work.  I write just about everything in python, as I said in the post.


 The point of the discussion is that I find most people equating
script with simple and hence by linking python with scripting
languages we may be doing python a bad turn.  it makes each 'sale' of
python harder since we have to fight against the mindset that feels
that since python is a scripting language, it cant do anything but toy
programs.

 Try to remember that the whole world is not populated with sysadmins.
Most of the world doesnt know what scripts can do, but they read about
Jscript and VBscript and see these as toy languages.  Or they are
introduced to perl or ksh from their admin friends and see someone
parsing a text file and figure thats all the language is good at.

 To a lot of language bigots out there, if it isnt compiled, or either
C or Fortran, it isnt appropriate.  Its caused by the era in which
they 'grew up' with computers I guess.  I fight daily with an older
engineer that refues to use anything but Fortran, even when it is
clearly not best suited for the problem.  I have nothing agains
Fortran, but in some cases, it doesnt make sense to use it.  

 Convincing them to use C/C++ is a lot easier than using Python
because C is at least similar to Fortran, and they seem to be more
comfortable with it.  Similar but not quite as extreme situations
arise all the time.  When I suggest rewriting an old Fortran or Cobol
or even C program in Python/C (sometimes C/C++ extensions are needed
to make things work) I get told that the risk is to high, but when I
suggest that it be re-done in just C or C++, no one says anything.  

By linking Python with scripting, (something I dont do, but they find
references suggesting this) they assume its a toy language that cant
do anything and in most cases wont even explore what it can do.  They
stop before they even get started.  and so when I suggest that we
could do something faster, easier, etc in python, they generally just
dismiss the idea until I have the chance to prove it to them.

Thats the point I was trying to make.  Not that python isnt a real
language.  I use it for everything except things where extremely high
performance is required (like drivers, embeded stuff on slow chips, or
very long scientific/mathematical simulations and calculations --  on
a 2 week simulation run, shaving 15% of the runtime of the app is very
meaningful, and the slow performance of parts of python are worth
avoiding).  In many cases, though, I wrap the C parts that do the high
performance work in to a pyhton module so I can still work on the
problem in python, but still get the performance of
C/C++/Fortran/WhatEver.  But with the title scripting language being
attached to python here and there, some people wont ever explore it
far enough to know that you can do this kind of high performance work
with it.  


-G




From anno4000 at lublin.zrz.tu-berlin.de  Sat Aug 14 23:22:08 1999
From: anno4000 at lublin.zrz.tu-berlin.de (Anno Siegel)
Date: 15 Aug 1999 03:22:08 -0000
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu>
Message-ID: <7p5bp0$v48$1@lublin.zrz.tu-berlin.de>

Tom Christiansen  <tchrist at mox.perl.com> wrote in comp.lang.perl.misc:

>I don't remember whether it also had these, but I have
>seen them elsewhere:
>
>    #define forever()  for(;;)
>    #define unless(expr) if(!expr)

Let's just hope they had the sense to really

     #define unless(expr) if(!(expr))

or else all bets are off.

Anno



From jcw at equi4.com  Sat Aug  7 20:12:58 1999
From: jcw at equi4.com (Jean-Claude Wippler)
Date: Sun, 08 Aug 1999 02:12:58 +0200
Subject: ANN: MkWrap - another MetaKit database extension
Message-ID: <37ACCB89.E9956900@equi4.com>

MetaKit is an embedded database engine, which can be used from Python
(sort of), Tcl (solid), and C++ (the native interface: rock solid).

MkWrap is a new "thin" interface from Python, which stays closer to the
core MetaKit API than the previous "Mk4py" interface.  It exposes far
more of the core funtionality, including selection, sorting, projection,
join, groupby, set operations, fast sequential searching, and regexp-
based matching on any string field.

As with the first Mk4py release, this is a proof-of-concept release,
intended to explore the effectiveness of such an interface in Python.

Caveat:  MetaKit doesn't support SQL, nor is it compliant with Python's
DBI - it lives in the twilight zone between flat files, OO databases,
relational systems, and tree-structured storage.

Keywords which *do* apply to MetaKit are: portable, high-performance,
low footprint, embeddable, transaction-based, dynamic schemas, multi-
language, load-on-demand, unconventional.

For details on this first alpha release of MkWrap, see:
        http://www.equi4.com/metakit/mk4py/mkwrap.txt
It contains pre-built extensions for Windows and Linux, a small sample,
rudimentary documentation, and the C++ sources of the Python interface.

MkWrap was initially created by Gordon McMillan - I merely sprinkled 
a few extra ideas and bugs onto it to make life more interesting :)

The MetaKit home page is here:
        http://www.equi4.com/metakit/

The Mk4py version of the Python interface is unaltered:
        http://www.equi4.com/metakit/mk4py/

Some background info on MetaKit can be found on my personal pages:
        http://www.equi4.com/jcw/

Comments, suggestions, problem reports always welcome (but see below).

-- Jean-Claude

P.S. This is a "deadline-crush" announcement - I'll be back August 18.



From 71351.356 at compuserve.com  Tue Aug 17 22:53:47 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Tue, 17 Aug 1999 22:53:47 -0400
Subject: Wacky Programming Tales (Was Re: Why use Perl when we've got Python?!)
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ced8@cs.colorado.edu> <slrn7r9lnb.pu8.sholden@pgrad.cs.usyd.edu.au> <slrn7r9nvr.h3c.jstevens@bamboo.verinet.com> <slrn7r9qu0.pu8.sholden@pgrad.cs.usyd.edu.au> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net>
Message-ID: <7pd7f9$ais$1@ssauraaa-i-1.production.compuserve.com>

That's not so strange.   Perhaps at a software house with a lot of propeller
heads, but at a place where application customization and other 'light
projects are done, little things like that help those who do not dream of
algorithms in their spare time or who don't plan to specialize and master a
particular language.   Besides once you've been bitten by the '==' in C (and
you don't use a good lint, because the boss won't buy one), #define EQ ==,
is not a sin.   Well, yes it is, but its not, but.....


Brad Howes <bradh at mediaone.net> wrote in message
news:wjy7lmxbwh7.fsf_-_ at bradh.ne.mediaone.net...
>snips<

> Hey! I actually worked on a project where someone did this! Also in their
> funky include file was
>
>   #define AND &&
>   #define OR  ||
>
> It was the most bizarre thing I ever saw -- not so much the include file,
> but the resulting code. As if a Pascal moths had come in, eaten at some
> code and left little Wirth fragments sprinkled around. Has anyone else
> encountered such strange programming behavior?
>
> ...now back to my language is bigger/longer/wider/stiffer than yours
>
> --






From bmcd at es.co.nz  Sun Aug 15 18:18:59 1999
From: bmcd at es.co.nz (Ben Caradoc-Davies)
Date: 15 Aug 1999 22:18:59 GMT
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net> <slrn7re808.77t.kaz@ashi.FootPrints.net>
Message-ID: <slrn7refg6.d4.bmcd@ripley.sulaco.net>

On Sun, 15 Aug 1999 20:15:44 GMT, Kaz Kylheku <kaz at ashi.FootPrints.net> wrote:
[reinventing "for"]
>So what's wrong with that? Granted, the for loop construct would have
>been better for the task instead of while, and the expensive consultant should
>have seen that. But what about the i = i + 1 increment? It smacks of BASIC, but
>there is nothing wrong with it.  It's not any less efficient than i++ or ++i.

The question is not one of efficiency but one of expressing the intent of the
programmer. Seeing this written as a while loop made me double take ...
where's the trick? What is going on here?

There is nothing wrong with i = i + 1 (especially for us Pythonistas :-), but
taken together with the loop it suggests an unfamiliarity with the accepted
idioms of the language.

>Maybe the consultant is anally retentive about not using operators other than
>assignment that have the side effect of modifying an object, and not using more
>than one side effect in a full expression. That approach leads to uninsipiring
>code, but nevertheless code that won't possibly have undefined behaviors due to
>ambiguities.
>In other words, with his ultra-conservative approach, the weenie is unlikely
>to ever things like a[i] = i++.

I wish this were the case, but given the rest of the code, I doubt it. You are
however, most charitable and tolerant :-)

>I'm afraid you don't have a case, if the program otherwise fulfilled its
>functional, performance and reliability requirements.

[Warning: may contain traces of rants]

The program did have some annoying bugs, and it was in the hunt for these that
the, er, unusual code was found. However, the program did mostly fulfil its
functional, performance, and reliability requirements, as they existed at the
time. And here's the rub:

A program must fulfil more than just those requirements, but also that of
maintainability. I'm sure I don't have to remind anyone here that a piece of
source code is a living document, and should be amenable to change (which is
why I was working on it in the first place).

The real problem I had with the project in question was that (except for some
crucial headers), the entire source code was almost uncommented except for a
copyright notice at the top of each source file. Understanding the code was an
exercise in reverse-engineering the intent of the programmer. I've seen this in
scientific software (such as the 49000 LOC of uncommented Fortran 77 written by
an engineer with english as a second language, with no use of procedures and
the entire flow-control managed by circuitous GOTOs and global variables), but
in a business environment it is unforgiveable.

When I left, I passed the [C++] project on to my successor, who was quite a
greenhorn. Meanwhile, the original author had fled the country. Now the current
project manager ("ANSI C? No I haven't heard of that company.") is leaving too.
The only record of what the code is meant to do is some outdated manuals and
the source code. This is mission-critical business software.

>The guy programs for accounting firms. So it's a given that he can't possibly
>be a great hacker who writes slick code. He does things that some of us
>wouldn't touch with a ten foot pole.  His high cost could have something to do
>with his domain specialization, and with a high profit margin on the part of
>the contracting company (assuming he's not freelance).

I'm sure he would have only seen a tiny slice of what the multinational charged
for his time. However, I think his price was due more to the gullibility of the
client and wandering project control than the domain. This is a bog-standard
database front-end!

Phew!

-- 
Ben Caradoc-Davies <bmcd at es.co.nz>



From ahopkins at ahopkins.dynacare.com  Thu Aug 26 10:02:23 1999
From: ahopkins at ahopkins.dynacare.com (Albert Hopkins)
Date: 26 Aug 1999 14:02:23 GMT
Subject: db modules for Python/NT?
Message-ID: <slrn7sai2h.a0a.ahopkins@ahopkins.dynacare.com>

I don't have a copy to work with, but does anyone know offhand if the
binary Python distribution include any one of the db modules?



--
Albert Hopkins
Sr. Systems Specialist
Dynacare, Inc
ahopkins at dynacare.com




From S.I.Reynolds at cs.bham.ac.uk  Thu Aug 19 16:55:09 1999
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Thu, 19 Aug 1999 21:55:09 +0100
Subject: Tk's default Toplevel - counterintuitive
Message-ID: <37BC6F2D.30A7@cs.bham.ac.uk>

Does anyone know how to tell if Tk has already instaniated a Toplevel
window? I want to be able to open several Toplevel windows in any order.

Something like:


def makeNewWindow():
   if defaulttoplevel in use:
      mstr=Toplevel
   else:
      mstr=None                #Uses the default Toplevel

   f = Frame(master=mstr)
   ...



Cheers,

Stuart

PS. Wouldn't it have made more sense if using None as a window's master
caused a new Toplevel window to be created rather than the situation now
where you end up sticking items all in the same window. The following
gives 1 window containing 3 buttons (doh!):

---
from Tkinter import *

def makeNewWindow():
    f = Frame(master=None)
    b=Button(master=f, text="Hello")
    b.pack(side=LEFT, fill=BOTH)
    Pack.config(f)

makeNewWindow()
makeNewWindow()
makeNewWindow()

---

A naive alternative is:
---
def makeNewWindow():
    f = Frame( master=Toplevel() )
    ...as before
----

Which gives you three separate windows but a random extra empty Toplevel
window (the mystery default `None' toplevel)... DohDoh!



From m.faassen at vet.uu.nl  Sun Aug 22 18:37:53 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 22 Aug 1999 22:37:53 GMT
Subject: How to sell Python to your boss?
References: <ka8yaf7d84t.fsf@delightful.cs.hut.fi> <ka8vhaaz5z2.fsf@delightful.cs.hut.fi>
Message-ID: <7ppu41$bvm$1@newshost.accu.uu.nl>

Petri Mikael Kuittinen <eye at niksula.hut.fi> wrote:
> Well, it looks like it is Christmas. :-)

[Petri can use Python!]

> Thanks everyone who suggested good ideas to sell Python!

Congrats!

I was hired to do a quick hack (right now I'm implemented a website with Zope though), and I asked 'can I use Python?' and they didn't know what that was,
and said okay (it was a quick hack anyway). Now they're stuck with Python
in the form of Zope and a few other nice things. :) So I did a stealth
introduction, myself, sort of. Wonder what I'll have to try next time..

Regards,

Martijn




From tismer at appliedbiometrics.com  Fri Aug 20 17:42:00 1999
From: tismer at appliedbiometrics.com (Christian Tismer)
Date: Fri, 20 Aug 1999 21:42:00 GMT
Subject: making python man pages looks hard
References: <1103_935134343@shoal> <y0j3dxe436q.fsf@vier.idi.ntnu.no>
Message-ID: <37BDCBA8.216EC91@appliedbiometrics.com>


"Magnus L. Hetland" wrote:
> 
> connolly at w3.org (Dan Connolly) writes:
> 
> >
> > The source format is LaTeX. Converting from LaTeX is hard/messy:
> 
> Well - the docs aren't in standard LaTeX, are they? According to
> www.python.org/doc/doc:
> 
>   "With almost no basic TeX or LaTeX markup in use, [...] the markup
>   syntax is about the only evidence of LaTeX in the actual document
>   sources."
> 
> Since the format seems to be clearly and simply defined, it shouldn't
> be *that* difficult... Anyway - a transition to XML seems to be on its
> way, which will make all this trivial. (Actually, it wouldn't be that
> hard to go from the HTML versions either, I guess, with a little
> imagination. My point is - there is no reason to be spooked by the
> complexity of LaTeX here...)

In I:\cvsroot\python\dist\src\Doc\tools\sgmlconv
I find some README and a couple of scripts.
latex2esis.py
esis2sgml.py

and it looks to be able to produce XML. ?

Then I'd take an XML parser and try to generate man pages :-)
I have just no idea of *that* format.

ciao - chris     (starting a "who writes the shortest man.py")

-- 
Christian Tismer             :^)   <mailto:tismer at appliedbiometrics.com>
Applied Biometrics GmbH      :     Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101   :    *Starship* http://starship.python.net
10553 Berlin                 :     PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint       E182 71C7 1A9D 66E9 9D15  D3CC D4D7 93E2 1FAE F6DF
     we're tired of banana software - shipped green, ripens at home



From a.eyre at optichrome.com  Tue Aug 31 08:34:06 1999
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Tue, 31 Aug 1999 12:34:06 GMT
Subject: deadlock when two servers call each other - will threads help?
In-Reply-To: <199908142158.QAA24395@dolphin.mojam.com>
References: <199908142158.QAA24395@dolphin.mojam.com>
Message-ID: <000701bef3ad$1dca94e0$3acbd9c2@peridot.optichrome.com>

> This is fine in theory, but in practice it yields deadlocks.  A child can
be
> making an XML-RPC call to its parent to forward one change request at the
> same time the parent is trying to make another call to that child.  Both
> servers will block in read() waiting for a response that never arrives.

Is there a non-blocking read() call like the BSD-sockets-style select() ?

--------------------------------------------
Adrian Eyre <mailto:a.eyre at optichrome.com>
Optichrome Computer Solutions Ltd
Maybury Road, Woking, Surrey, GU21 5HX, UK
Tel: +44 1483 740 233  Fax: +44 1483 760 644
http://www.optichrome.com
--------------------------------------------




From bwarsaw at cnri.reston.va.us  Tue Aug 17 23:20:53 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Tue, 17 Aug 1999 23:20:53 -0400 (EDT)
Subject: Regarding the magical "or" operator
References: <37b5d712@cs.colorado.edu>
	<37B9D02C.A034B044@compaq.com>
Message-ID: <14266.9877.84844.115308@anthem.cnri.reston.va.us>

>>>>> "GE" == Greg Ewing <greg.ewing at compaq.com> writes:

    GE> Personally I think there should be a separate Boolean type,
    GE> and it should be an error to test the truth of any other type.

Last time the idea of a boolean type was brought up, I think we zeroed
<wink> in on the attached Python prototype.  I'm not sure I like the
second half of your proposal; it works okay I s'pose in Java, but
doesn't seem right in Python.

-Barry

-------------------- snip snip --------------------
class Bool:
    def __init__(self, flag=0):
        self.__flag = not not flag

    def __str__(self):
        return self.__flag and 'true' or 'false'

    def __repr__(self):
        return self.__str__()

    def __nonzero__(self):
        return self.__flag == 1

    def __cmp__(self, other):
        if (self.__flag and other) or (not self.__flag and not other):
            return 0
        else:
            return 1

    def __rcmp__(self, other):
        return -self.__cmp__(other)

true = Bool(1)
false = Bool()



From S.I.Reynolds at cs.bham.ac.uk  Thu Aug 19 11:30:13 1999
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Thu, 19 Aug 1999 16:30:13 +0100
Subject: Canvases not resizable?
Message-ID: <37BC2305.B3C@cs.bham.ac.uk>

Are Tkinter canvases fixed to one size? I don't seem to be able to get
them to expand to fill their container when its resized even the though
the canvas pack options are set to fill=BOTH  and expand=1.

The example both creates a Frame with a canvas. Resizing the frame
doesn't change the size of the canvas.

Stuart


---------

from Tkinter import *



class CFrame(Frame):
    """A Frame containing a Canvas
    """
    def __init__(self, w, h):
	Frame.__init__(self, master=None)

	#Set the frame's width and height
	self["width"] = w
	self["height"] = h

	#Create a red canvas
	self.canvas = Canvas(master=self, bg="red")

	#Print canvas dimensions when clicked
	self.canvas.bind("<Button-1>", self.printCanvasSize)

	#These should make the canvas fill the frame right??
	self.canvas.pack(fill=BOTH, expand=1)


    def printCanvasSize(self, point):
	print 'w = ', self.canvas["width"]
	print 'h = ', self.canvas["height"]


if __name__ == '__main__':
    
    cframe = CFrame(w="8i", h="8i")
    Pack.config(cframe)
    cframe.mainloop()



From philh at vision25.demon.co.uk  Sat Aug 21 05:19:26 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sat, 21 Aug 99 09:19:26 GMT
Subject: can some one help a newbie?
References: <vdgv3.13036$x04.783764@typ11.nn.bcandid.com> <7pkke6$lut@dfw-ixnews6.ix.netcom.com>
Message-ID: <935227166snz@vision25.demon.co.uk>

In article <7pkke6$lut at dfw-ixnews6.ix.netcom.com>
           aahz at netcom.com "Aahz Maruch" writes:
> Overall, if your background is as a sysadmin using shell scripts, grep,
> awk, sed, and so on, you'll find Perl a bit easier.

I've used all these, and find Python easier than Perl. Perl would
probably be easier than Python for 1-liners, if I could remember
the syntax.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From tchrist at jhereg.perl.com  Sat Aug 14 16:05:43 1999
From: tchrist at jhereg.perl.com (Tom Christiansen)
Date: Sat, 14 Aug 1999 20:05:43 GMT
Subject: Why use comp.lang.python when you've got comp.lang.perl.misc? 
In-Reply-To: Christian Tismer's message of Sat, 14 Aug 1999 22:03:33 +0200 .
             <37B5CB95.3FCF571F@appliedbiometrics.com> 
References: <37B5CB95.3FCF571F@appliedbiometrics.com>
Message-ID: <199908142005.OAA23215@jhereg.perl.com>

After reading through the _Programming Python_ book, I thought to
summarize some of my observatations, but the current climate is just
too hot.  I liked the "Python for C++" people at the back of the book.
We have something a little like this for Perl in the perltrap manpage.
Maybe later when the world is quieter.

--tom



From sjmachin at lexicon.net  Tue Aug 31 08:36:10 1999
From: sjmachin at lexicon.net (John Machin)
Date: Tue, 31 Aug 1999 12:36:10 GMT
Subject: Soundex implementation (was: RE: Fuzzy string matching?)
In-Reply-To: <000001bef12c$139c0680$ad2d153f@tim>
References: <000001bef12c$139c0680$ad2d153f@tim>
Message-ID: <19990831123108417.AAA352.267@max41072.izone.net.au>

A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 1784 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/19990831/085febcf/attachment.bin>

From hei at adtranzsig.de  Wed Aug 11 18:15:12 1999
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 12 Aug 1999 00:15:12 +0200
Subject: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <q43s3.58867$jl.37270547@newscontent-01.sprint.ca>
Message-ID: <934412029.374060@news.adtranzsig.de>

Bruce Dodson schrieb in Nachricht ...
>Ni!  I don't see the point of creating a variant of python that aims to be
>more efficient, but does so at the cost of taking away Python's most
>powerful optimization mechanism: external C modules.

Here, i agree with Bruce.
At the moment, it's common practice to accelerate a Python
program that's too slow by moving parts of it into homebrew
extension modules. Taking away this option would maybe
invalidate Viper as a means of optimization. It would be okay
with me, though, if you allow a different kind of extension
modules (a different interface to the extension module)-
if ViperC creates C code, i don't see a big problem with
that.

>I also think that a
>python that has enough knowledge of types to perform inlining... well, such
>a beast is no longer Python, no matter how similar the syntax appears.

In this case, i disagree with Bruce. Implementing a subset
of Pythons semantic possibilities and keeping the same syntax makes
sense in real life. Just try to write your production code in a less
wild fashion than usually, not changing bindings all over, and you'd have
a comfortable way of getting at more speed without rewriting as
much stuff as if you had to port it to C.

Dirk







From bryn at mds01.itc.com.au  Sat Aug  7 04:33:13 1999
From: bryn at mds01.itc.com.au (B Kingsford)
Date: Sat, 7 Aug 1999 18:33:13 +1000
Subject: Python, Embedding Q's
Message-ID: <7ogr96$e2g$1@wyrm.its.uow.edu.au>

hi all

im embedding Python in multi-threaded server, and just had a
couple of q's regarding compiling files, cache, etc. if any help
or thoughts could be offered, please do.

I wish to use Python's built-in threading of the interpreter,
aswell as Python's memory management [for portability], will
this have any major, adverse performance effects?
Does Python handle which thread does what?
would it be better to create say, x threads, each with their
own instance of the interpreter?

what is the PyDict format of globals/locals from PyModule_GetDict?
what vars are declared in this ModuleDict? is it necessary to add
anything there?

does PyCodeObject keep (or represent) the compiled code in mem?
would it be worth maintaining a heap of PyCodeObjects (one for
each .pyx, or until cache is full...) as opposed to recompiling
for each request (unless it is desired, say, .pyd), say
PyCodeObject  * pcoCached[] = malloc( 500000 );
will Py use this new heap for the CodeObject pointers aswell as?
will the CodeObjects fill the heap, or will Py, when compiling
for a CodeObject, create mem elsewhere? eer, does
sizeof(PyCodeObject) include the Py bytecode?

if i were to maintain the PyCodeObjects, how then would i have
an instance of the interpreter run this code? ahh, obviously
PyEval_EvalCode, but what if Python is handling the sub-interpreters?
how to say, ThatSubInterpreter->EvalCode( PyCodeObject )?

is Py_NewInterpreter really a 'fresh copy' of the main interpreter
for use by a thread? if i add built-in mods with PyImport_AddModule,
do i need to do this for each Py_NewInterpreter? what about
Py_InitModule? does this need to be called for each new aswell?

in the examples, the telnet server seems to call Py_NewInterpreter
for each request. would i have to, if say i wished to limit the
worker threads, when a request occurs, if threads >= max, put
request in queue, OR, create a new thread with Py_NewInterpreter,
handle the request, check to see if any new requests are in the
queue [and handle them too], then Py_EndInterpreter? what about
reusing Py_NewInterpreters after they've been idle for a while?
(say, a LIFO q of interpreters... you know how it is... no action
for a while then bang, everybody wants in on your server!!)

cheers and thanks in advance
bryn





From jblaine at shell2.shore.net  Sun Aug 15 16:54:32 1999
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Sun, 15 Aug 1999 20:54:32 GMT
Subject: C's isprint() concept?
Message-ID: <slrn7rea88.9uf.jblaine@shell2.shore.net>

If I want to replace all non-printable characters in a string with a single
space, what would be the best way?  Do I need to loop over the entire string
character by character checking the ord() value of each one?  Anyone have
a sane way to do this with regular expressions?

[ C has isprint() and isgraph() macros ]

Thanks for any help



From greg.ewing at compaq.com  Mon Aug 23 00:25:12 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Mon, 23 Aug 1999 16:25:12 +1200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <37BC1ED4.A29883BC@appliedbiometrics.com> <37bc2588@cs.colorado.edu>
Message-ID: <37C0CD28.1B37A0FB@compaq.com>

Tom Christiansen wrote:
> 
> As for the tuple thing, why can I say
> 
>     print "You have $%0.2f bucks" % (4.5/100)
> 
> How is that a singleton tuple?  Don't I have to use
> a trailing comma?

The formatting operator recognises a special case:
if you give it something which isn't a tuple, it
coerces it into a 1-tuple for you. This saves you
from having to use the awkward 1-tuple syntax when 
the format string has only one % item in it.

>  It won't let me!  Why is
> 
>     print "You have $%0.2f bucks" % (4.5/100,)
> 
> illegal,

It's not illegal! It works just fine:

>>> print "You have $%0.2f bucks" % (4.5/100,)
You have $0.04 bucks

> but
> 
>     print "You have $%0.2f bucks" % ((4.5/100,) * 2)
> 
> mandatory?

No, that one doesn't work:

>>> print "You have $%0.2f bucks" % ((4.5/100,) * 2)
Traceback (innermost last):
  File "<pyshell#9>", line 1, in ?
    print "You have $%0.2f bucks" % ((4.5/100,) * 2)
TypeError: not all arguments converted

> Was this (elt,) thing just an, er, unforeseen design
> misfeature?

For what it's worth, I don't much like it either. It would
be better if tuple construction had a clear and unambiguous
syntax like list and dictionary construction. But there don't
seem to be enough kinds of brackets left over in the ascii
character set...

Greg



From aboukirev at iname.com  Sun Aug 29 18:39:53 1999
From: aboukirev at iname.com (Alexei Boukirev)
Date: Sun, 29 Aug 1999 17:39:53 -0500
Subject: Why do we call python a scripting language?
References: <37c6bb11.340373891@midpoint> <37c66347.518331@news.netset.com> <7q6imd$kuo$1@nntp6.u.washington.edu>
Message-ID: <RRiy3.28518$eW1.5110@nntp0.chicago.il.ameritech.net>

Donn Cave <donn at u.washington.edu> wrote in message
news:7q6imd$kuo$1 at nntp6.u.washington.edu...
> Quoth tdixon.no at spam.fwi.com (Timothy Dixon):
> | On Fri, 27 Aug 1999 10:18:55 GMT, guy_oliver at yahoo.com wrote:
> |> Why do we call python a scripting language?
> |>
> |> For me, and for most of the people I try to tell about the benefits of
> |> python, the title 'scripting language' implies that its not a real
> |> language capable of real programs.  It implies to most of them that it
> |> is good for simple sysadmin and maybe short cgi scripts.
> |
> | I put it in the class of scripting languages because it's embeddable
> | (as I understand it; I'm actually just a newbie) in other applications
> | where it can be used for scripting another app.
>
> Yes!  This is the useful meaning for "scripting", as I see it!  The
> definitions that try to distinguish between ``quick, inconsequential''
> vs. ``serious, production'' etc. are hopelessly lame, regardless of
> their distinguished sources, because it's a useless struggle to assign
> importance this way.  A 3 line Bourne shell script can be mighty
> important production software.
>
My understanding of scripting (vs. interpreted) language is that for
scripting language byte compiler (or precompiler if you wish) is an integral
part of run-time environment.  So you effectively can load source code at
run-time and execute it.  This holds true for many languages, including
various Lisp/Schema implementations, even Smalltalk (except for end products
stripped off original environment).  Not true for Java as it does not allow
to distribute bite-compiler related classes (part of JDK, not JRE), and even
then this integration there is not well documented.

Funny how all Emacsers call Elisp a 'programming' language and not
'scripting' language (though it IS scripting).  Is it because it has that
huge development support environment which is being considered 'OS' by
itself (and Elisp is not used outside of it)?

Same would probably hold true for Java once JavaOS is complete and
operational.

Shell is on the other hand is still considered 'scripting' environment.

May be it's a matter of preference how you call it?

One more explanation of 'scripting' language could be that any program
written in it requires some environment to run in which is not simply a
language interpreter but more like a complete usable application itself.

Alexei





From philh at vision25.demon.co.uk  Thu Aug 12 16:10:53 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Thu, 12 Aug 99 20:10:53 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk> <luemh851ym.fsf@stest.ch>
Message-ID: <934488653snz@vision25.demon.co.uk>

In article <luemh851ym.fsf at stest.ch>
           christian.egli at alcatel.ch "Christian Egli" writes:
> Paul> Glade is a GUI application that allows you to build a window
> Paul> visually.
>  
> Phil> Okay, but I saw a reference somewhere to Glade having a
> Phil> declarative file format that could be converted into code to
> Phil> build the GUI. I was mostly talking about the declarative file
> Phil> format.
> 
> Maybe you meant the storage format of Glade which is XML which in turn
> can be used with Libglade to build the GUI dynamically.

Yep. Read the docs today.
 

-- 
Phil Hunt....philh at vision25.demon.co.uk




From billtut at microsoft.com  Tue Aug 10 20:00:09 1999
From: billtut at microsoft.com (Bill Tutt)
Date: Wed, 11 Aug 1999 00:00:09 GMT
Subject: Getting at line numbers with parsermodule
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8100DB91066@RED-MSG-50>

> From: Charles G Waldman [mailto:cgw at fnal.gov]
> 

> What I'd like to be able to do is, given any node in an AST object
> returned by compile.suite, is to know the line number in the input
> that corresponds to this node.  Is this doable?  Any suggestions?
> 

You could just use parser.ast2tuple(ast, 1)
The extra parameter adds the line # as the last element of each AST node
tuple.

Bill



From pj at sgi.com  Sat Aug 21 14:18:02 1999
From: pj at sgi.com (Paul Jackson)
Date: 21 Aug 1999 18:18:02 GMT
Subject: Man pages would be cool.
References: <37BC6986.A66B044E@starmedia.net> <37BE24E0.79EA064E@montana.com> <37BE9EAB.CB7B1380@prescod.net> <37BEDBF3.BCA64E58@montana.com>
Message-ID: <7pmqgq$354ai@fido.engr.sgi.com>

bowman wrote:
|> a readily apparent escape hatch ... unlike ... emacs ... pre-menu

Or even 'vi'.  I recall for 6 months when I was bringing up a
68010 Unix box, I'd just reboot the lab machine anytime I found
it left in vi by another engineer, because I didn't know how to
quit vi.

-- 

=======================================================================
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj



From jblaine at shell2.shore.net  Fri Aug 20 14:06:45 1999
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Fri, 20 Aug 1999 18:06:45 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <37bc313b@cs.colorado.edu> <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us> <019301beeaea$004c4d80$f29b12c2@secret.pythonware.com> <37bd4876@cs.colorado.edu>
Message-ID: <slrn7rr69l.ata.jblaine@shell2.shore.net>

>It's a shame that so many of you should be so unbelievably harsh on me
>when *I'm* the one who with malice toward none and full of revulsion
>at the flames causes by *YOUR* community's ignorant attacks on mine,
>that *I* made the honest effort to learn *YOUR* language, and that I

I can understand your frustration, but also kinda feel you did your share
of stooping to the flamewar mentality of one-off flippant responses to
fuel this thread's more fired-up posts.

>furthermore for your benefit not my own then took the additonal trouble
>to write-up my experiences and observations gleaned while five hundred
>miles away from the nearest tree or internet link.

I doubt a Python ambassador doing the same as you did and then heading over
to comp.lang.perl.misc would have been met any more kindly.  I personally
think a disclaimer-ish statement at the top of your posting which explained
what the POINT of your posting was would have helped drastically.

But that doesn't make things right, and I for one was pretty stunned at a
lot of the responses generated by the Pythoners in this newsgroup.  I had
never seen such hostility in this newsgroup before.  But Tom, you definitely
appear to be focusing on the half of the people who got pissy.  Many people
stated they were thankful, glad, whatever and many others took their
precious time (just like yours) to go through your list and try to clear
some items up, let you know that there was work being done on XYZ, or flat out
tell you "Yes, foo in Python sucks."

>But no, that's not it.  It's just Usenet, the place where condescension,
>conceit, and cruelty all come far too easily.  The more the pissier,

That has definitely not been my experience in *this* newsgroup in the
last 2 years.  Many others, but not this one.



From MHammond at skippinet.com.au  Thu Aug 26 01:57:53 1999
From: MHammond at skippinet.com.au (Mark Hammond)
Date: Thu, 26 Aug 1999 15:57:53 +1000
Subject: Python Sighting
References: <slrn7rlrsu.l2e.mryan@unix2.netaxs.com>
Message-ID: <7q2lgh$aei$1@m2.c2.telstra-mm.net.au>

Michael W. Ryan wrote in message ...
>I made an interesting discovery the other day while roaming through the
>registry of an NT system trying to troubleshoot a problem in a Microsoft
>Site Server class I was supporting:  Microsoft Site Server 3.0 (not sure
>if it's the normal or commerce edition) uses Python.  From the presence of
>the ni.pyc file, I'm going to guess that it's 1.4.x.  I'm also not sure of
>what capacity they're using it in.

And we indirectly have that MS package to thank for win32com :-)  If you
search deja-news, you should find a number of references to this.

Mark.






From ajung at sz-sb.de  Thu Aug 26 16:16:36 1999
From: ajung at sz-sb.de (Andreas Jung)
Date: Thu, 26 Aug 1999 20:16:36 GMT
Subject: Python and Oracle
In-Reply-To: <37BA3813.2BEB45A8@netscape.com>; from vijay Mallela on Tue, Aug 17, 1999 at 09:35:31PM -0700
References: <37BA3813.2BEB45A8@netscape.com>
Message-ID: <19990826221636.A21122@sz-sb.de>

On Tue, Aug 17, 1999 at 09:35:31PM -0700, vijay Mallela wrote:
> Is there any current work going on with the Oracle module ?
> Thanks

Take a look at DCOracle - this should work with Oracle 7/8:

	http://www.zope.org/Download/DCOracle


-- 
                                _\\|//_
                               (' O-O ')
------------------------------ooO-(_)-Ooo--------------------------------------
   Andreas Jung, Saarbr?cker Zeitung Verlag und Druckerei GmbH
   Saarbr?cker Daten-Innovations-Center
   Untert?rkheimerstra?e, D-66103 Saarbr?cken, Germany
   Phone: +49-(0)681-502-1528, Fax: +49-(0)681-502-1509
   Email: ajung at sz-sb.de (PGP key available)
-------------------------------------------------------------------------------



From kernr at mail.ncifcrf.gov  Tue Aug 10 22:05:41 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Wed, 11 Aug 1999 02:05:41 GMT
Subject: Compiling Python Extensions on Win32 with GCC 2.95
Message-ID: <37b0da65.5015249@news.erols.com>

With the recent release of GCC 2.95, I have updated my instructions
for compiling Python extensions on Windows 95/98 and NT systems using
free tools.  My previous instructions explained the setup for using
EGCS 1.1.2, an "experimental" version of GCC.  Since EGCS merged into
GCC to make GCC 2.95, the instructions could be updated to use this
newest compiler.

To provide some background, GCC (GNU Compiler Collection) is a GPL'ed,
well, compiler collection that includes front ends for C, C++,
Fortran77 (with some optional Fortran90 constructs, IIRC),
Objective-C, and (now) a few, experimental Chill and Java front ends.
More information on GCC 2.95 is available at 
<URL: http://egcs.cygnus.com>.

It is possible now to compile programs for the i386-mingw32 target, a
native target for Win32 systems.  Such programs use a Microsoft DLL as
its C runtime library (CRTDLL.dll by default, MSVCRT.dll as an
option).  Unlike the i586-cygwin32 target 
<URL: http://sourceware.cygnus.com/cygwin>, the i386-mingw32 target
does not require the program to link with the Cygwin DLL.
Consequently, there are *no* licence restrictions imposed on the
resulting program (not by the compiler at any rate; linking against
third-party code might restrict use).  Information on the Win32 port
of GCC, and the port itself, are at 
<URL: http://www.xraylith.wisc.edu/~khan/software/gnu-win32/gcc.html>.

Paul Sokolovsky originally patched the Python headers to allow the
then-EGCS to compile Python extension modules for the i386-mingw32
target.  I now maintain a web page containing instructions on setting
up the compiler, headers, and the other related paraphenalia required.

My page is 
<URL: http://starship.python.net/crew/kernr/mingw32/Notes.html>.

<P><A
HREF="http://starship.python.net/crew/kernr/mingw32/Notes.html">Compiling
Python Extensions on Win32 with GCC 2.95</A> - Instructions to assist
compiling Python extensions on Win32 OS's with free tools (10-Aug-99)

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From tim_one at email.msn.com  Wed Aug 11 02:45:45 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Wed, 11 Aug 1999 02:45:45 -0400
Subject: Viper:  yet another python implementation
In-Reply-To: <q43s3.58867$jl.37270547@newscontent-01.sprint.ca>
Message-ID: <000101bee3c5$2393bde0$8d2d2399@tim>

[Bruce Dodson]
> ...
> I also think that a python that has enough knowledge of types to
> perform inlining... well, such a beast is no longer Python, no matter
> how similar the syntax appears.

Simple inlining doesn't require any knowledge of types, and would be
valuable in many programs.  The thing that stops it today is that you just
mever know what a name is bound *to*; e.g.,

def f(i):
    return i

def g(i):
    return i+1

def h(indices):
    sum = SomeType()
    for i in indices:
        sum = sum + f(i) + g(i)
    return sum

For all Python knows, the invocation of f, the invocation of g, the hidden
invocation of indices.__getitem__, and/or the invocation of f(i).__radd__ or
sum.__add__ or (sum + f(i)).__add__ or g(i).__radd__ may reach into the
module globals and change the bindings of "f", "g", or both, to anything at
all, or even to different things on each loop iteration.

This is great when you need it, but is likely of no use to 99.9999% of
runtime Python cycles.  If John freezes the module bindings, the guts of the
loop can be mindlessly transformed into

       sum = sum + i + (i+1)

regardless of types with no change in semantics, a major boost in speed,
and-- if done straightforwardly --a major loss of debugability when an add
raises an exception.

there-ain't-no-free-lunch-but-some-good-meals-are-cheap-ly y'rs  - tim





From paul at prescod.net  Thu Aug 26 10:33:40 1999
From: paul at prescod.net (Paul Prescod)
Date: Thu, 26 Aug 1999 14:33:40 GMT
Subject: Returning none
Message-ID: <37C55044.8CA02530@prescod.net>

I think it would be more pythonish if this would return a runtime (or
even compiletime) error:

def foo():
	a=5

b=foo()

It shouldn't just return None. That's a source of errors.

This is obviously an incompatible change so I propose it for Python 2.

 Paul Prescod



From sdm7g at Virginia.EDU  Mon Aug  9 16:50:29 1999
From: sdm7g at Virginia.EDU (Steven D. Majewski)
Date: Mon, 9 Aug 1999 16:50:29 -0400 (EDT)
Subject: Packing and unpacking IP addresses into 32-bit integers
In-Reply-To: <ytt1zdcafmg.fsf@gilgamesh.cse.ucsc.edu>
References: <ytt1zdcafmg.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <Pine.A32.3.90.990809163915.17213C-100000@elvis.med.Virginia.EDU>

On 9 Aug 1999, Ben Gertzfield wrote:

> I noticed that the 'socket' module in Python has slightly different
> behavior for gethostbyname() than I'm used to.
> 
> In Python, gethostbyname() returns a string, not a packed IP address
> (a 32-bit integer) of the IP address, as C and Perl do.  However, I
> need to talk with C over the network, and the 'struct' module doesn't
> deal with packing or unpacking IP addresses.
> 
> Do I have to use a regular expression to split up the dotted quads
> that gethostbyname() returns into the component bytes and then use
> binary shifts and or'ing to get them into a standard 32-bit integer
> format?  Or is there a better way that I'm totally missing?
> 

No you don't have to use regular expressions and binary shifts and or's:

 First, map int or string.atoi onto the result of applying string.split
 to the return string of gethostbyname -- that will give you a list of ints.

   adr = map( string.atoi, string.split( gethostbyname( name ), '.' ))

and use either struct.pack or array to coerce it into a binary string:

  a = apply( struct.pack, [ 'bbbb' ] + adr )

(or longer, but less cryptically)

  a = array( 'b') 
  a.fromlist( adr )
  a.tostring() 
  a 



---|  Steven D. Majewski   (804-982-0831)  <sdm7g at Virginia.EDU>  |---
---|  Department of Molecular Physiology and Biological Physics  |---
---|  University of Virginia             Health Sciences Center  |---
---|  P.O. Box 10011            Charlottesville, VA  22906-0011  |---




From skip at mojam.com  Wed Aug 18 13:09:52 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 18 Aug 1999 12:09:52 -0500 (CDT)
Subject: representation of unbound methods - strange behaviour
In-Reply-To: <37bbecf1.100354231@news.omnilink.de>
References: <37bbecf1.100354231@news.omnilink.de>
Message-ID: <14266.59462.274125.746880@dolphin.mojam.com>

>>>>> "Stefan" == Stefan Franke <spamfranke at bigfoot.de> writes:

    Stefan> Why does this happen:

    ...

    <unbound method A.f> stuff snipped

I believe it is constructing the method name on-the-fly using the names it
finds in the class and function.  It's not using the attribute name.

    >>> class A: pass
    ... 
    >>> def foo(): pass
    ... 
    >>> A.f = foo
    >>> A.g = foo
    >>> A.f
    <unbound method A.foo>
    >>> A.g
    <unbound method A.foo>

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From aaron_watters at my-deja.com  Sat Aug 28 20:32:30 1999
From: aaron_watters at my-deja.com (aaron_watters at my-deja.com)
Date: Sun, 29 Aug 1999 00:32:30 GMT
Subject: Larry's and Guido's REAL sigs
Message-ID: <7q9v2u$1jk$1@nnrp1.deja.com>

At the ORA Open Source Conference I got
Mark Lutz and David Ascher to sign their
"Learning Python", but for fun I also got
Tom Christiansen, Larry Wall, and Guido
van Rossum to sign it too.

If you want to get the other sigs you'll
have to visit me or buy the book from me
(at a highly inflated price) but I make
public the page that Guido and Larry signed:

   http://www.chordate.com/sigs.GIF

enjoy!  -- Aaron Watters

===

How many Mac developers does it take to
change a lightbulb?
One.  she holds the bulb up and expects
the universe to revolve around her.
   --Guy Kawasaki (former chief Apple Evangelist)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From cribloco at aol.com  Tue Aug  3 09:06:21 1999
From: cribloco at aol.com (LOCOS CRIBMASTER)
Date: Tue, 03 Aug 1999 13:06:21 GMT
Subject: Earn thousands from your home! Do not IGNORE this message! EXTRA CASH.doc (1/1)
Message-ID: <hPBp3.5550$DM5.50974@newse2.tampabay.rr.com>

------------------Rover042856
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

If you want to make more money than you can imagine!
All you have to do is open the attatchment and check it out!
Total cost $8.00! Yes its TRUE! Check it out!

------------------Rover042856
Content-type: application/octet-stream; name="EXTRA CASH.doc"
Content-disposition: attachment; filename="EXTRA CASH.doc"
Content-Transfer-Encoding: base64
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=
------------------Rover042856--



From stephan at pcrm.win.tue.nl  Tue Aug 24 03:16:56 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 24 Aug 1999 09:16:56 +0200
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan>
Message-ID: <m3hflp4phz.fsf@pcrm.win.tue.nl>

forcer <forcer at mindless.com> writes:

> Hi there.
> 
> Lately, i've been thinking about Python. While it is an
> exceptionally clean language, it has one very obvious wart:
> The ``pass'' statement.
> 
> To quote the reference manual:
> 
>    pass is a null operation -- when it is executed, nothing
>    happens. It is useful as a placeholder when a statement is
>    required syntactically, but no code needs to be executed, for
>    example:
> 
>    def f(arg): pass    # a function that does nothing (yet)
> 
> So it happens that Python has places where semantically nothing
> should be done, while syntactically it requires a statement.

I don't agree. It is very important for a programming language
to have a "do nothing"-statement, if only to satisfy the
computer scientists among us. The first program (well, actually
Hoare-triple) we had to prove in first-year CS was:

{P} skip {P}

,were skip was the "do nothing"-statement in the Guarded Command
Language. It's essential to have such a statement, because
it behaves as the identity of the "composition" operator on
the statements in your language, i.e. you have that

pass; X

is equivalent to just

X

for every statement X. This is also true in Python.

Greetings,

Stephan



From news at dorb.com  Fri Aug 13 16:59:00 1999
From: news at dorb.com (Darrell)
Date: Fri, 13 Aug 1999 16:59:00 -0400
Subject: Question about order in lists
References: <7p1v28$vcf$1@news04.btx.dtag.de>
Message-ID: <DH%s3.206$O02.12373@typhoon.nyroc.rr.com>

Worked for me.

>>> l1=[5,3,7]
>>> l2=[8,6,3]
>>> x=[]
>>> for i in range(len(l1)):
...     x.append(l1[i]+l2[i])
...
>>> x
[13, 9, 10]

--
--Darrell

Thorsten Z?ller <tzoeller at t-online.de> wrote in message
news:7p1v28$vcf$1 at news04.btx.dtag.de...
> Suppose you write a code as follows:
>
> x = []                                                # empty list
> list1 = [5, 3, 7]                                  # list containing any
> numbers
> list2 = [8, 6, 3]                                  # another list
containing
> any numbers
> for i in range(len(list1)):
>     x.append(list1[i-1] + list2[i-1])       # [5+8, 3+6, 7+3]
> ...
> x                                                        # print resulting
> list
>
> I'm trying to achieve that, having two lists of the same lenght containig
> any numbers, each component of the two lists is added to his pendant of
the
> other lists (i.e. the first component of list1 is added to the first
> component of list2 etc). The result is saved in another list x. But if I
run
> this code, I do not get [13, 9, 19] for x, but [10, 13, 9], i.e. the
> components are in a wrong order now. Why does the order change and how can
I
> avoid it?
>
> Thank you
> Thorsten
>
>





From jsouto at dnet.br  Mon Aug  9 16:28:16 1999
From: jsouto at dnet.br (Jairo Souto)
Date: Mon, 9 Aug 1999 20:28:16 GMT
Subject: report generation
Message-ID: <m11Dw1o-0006P6C@iflg>

I couldn't find a module for [plain-text] report generation
from generic data (sql data-base, lists, etc..).

Is there such a module?

--Jairo Souto  jsouto at dnet.br  (+55) (38) 9971-0014




From befletch at my-deja.com  Sun Aug  8 15:13:55 1999
From: befletch at my-deja.com (Bruce Fletcher)
Date: Sun, 08 Aug 1999 19:13:55 GMT
Subject: Parsing HTML tags with 're'
Message-ID: <7okkte$25a$1@nnrp1.deja.com>

I'm trying to parse HTML using Python's 're' module.  Generally things
are working fine, but I have one small problem left, and I'd also like to
know if I'm just going the wrong direction entirely.

I have a set of expressions I use for different stages of my parsing
process.  These include:

TagStart   = re.compile( '<[a-zA-Z]|<!--|</[a-zA-Z]' )
TagEnd     = re.compile( '>' )
CommentEnd = re.compile( '-->' )

I use these to find the start and end of tags.  Since I parse line by
line, I can't assume the tag will end in the same string as it started
in, so I don't look for '<*>' or anything similar.  I look for a tag
start, and then I buffer lines until I find a matching tag end.  Once I
have a complete tag, I want to parse out the tag type and the argument
list where appropriate (i.e. not in comments, and end tags don't have
arguments).  To do this parsing, I use these expressions:

Structure = re.compile('^<(?P<type>[a-zA-Z_/]\w*)\s*(?P<args>[^>]+)*>')
Arglist   = re.compile('(?P<name>[^=]+)=?(?P<value>.+)?')

The first one separates out the tag type and the optional argument list,
and the second one parses out the individual arguments.

The only problem I'm having is with the arglist parser; if an argument
looks like, arg="quoted arg with spaces", the spaces cause the argument
to break up.  What's the best way to fix this?

As I said before, I would also be interested to know if there is a much
better way to do this.  Did I miss a standard HTML parsing Python module?
If this approach is reasonable, feel free to scoop the code for your own
nefarious purposes.

- Bruce


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From andrew at starmedia.net  Mon Aug 23 20:47:54 1999
From: andrew at starmedia.net (Andrew Csillag)
Date: Tue, 24 Aug 1999 00:47:54 GMT
Subject: "def" vs "sub" (was RE: More random python observations from a perl 
 programmer)
References: <001601beeaef$e795f380$712d2399@tim> <23081999.1@sanctum.jae>
Message-ID: <37C1EBBA.573DA45C@starmedia.net>

"Juergen A. Erhard" wrote:
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> >>>>> "Tim" == Tim Peters <tim_one at email.msn.com> writes:
> 
> [SNIPPED lots of interesting stuff]
> 
>     Tim> You have to grab a copy of PyLint for that (but then you are
>     Tim> a fan of separate tools, right <wink>?).
> 
> And could you tell us where this mythical PyLint is?  More advtising
> couldn't hurt (I have browsed python.org a coupl times, but have seen
> no mention... though I haven't looked for PyLint (didn't know it
> existed))
> 
> [SNIPPED remainder of mail]
> 
> Bye, J
> 
> PS: Or is this really "mythical" in that it only exists in Tim's
> imagination?  ;-)

Nope, it's a creation of the venerable Andrew Kuchling as part of the
kwParsing package (a parser generator package), which also includes the
gadfly relational database - way cool.

http://www.chordate.com/kwParsing/index.html
-- 
import string,sys;t=string;f=t.index;j=t.join;print(lambda
s:j(map(lambda
x,s=s:x[:f(x,':')+2]+j(map(lambda y:'%02X'%ord(y),x[f(x,':')+2:]),'
'),map
(lambda i,s=s:("%04X:
"%i)+s[i:i+16],range(0,len(s),16))),'\n'))(open(sys.
argv[1]).read())



From nick at #spam_me_not#videosystem.co.uk  Tue Aug 31 08:55:20 1999
From: nick at #spam_me_not#videosystem.co.uk (Nick Trout)
Date: Tue, 31 Aug 1999 13:55:20 +0100
Subject: RAD
References: <7qfsml$to8$1@news.is.co.za>
Message-ID: <37cbcee8@news.xtml.co.uk>

| I am thinking of doing a Development enviroment for python as a way of
| learning python and just for the heck of it.
|
| My question is > How would you programaticaly find the return type , and
the
| types of the parameters of a method / function . How do I find the methods
| of a
| class programmatically.

dir(<classname>) returns a list of method names including the builtin ones.
eg. __docs__

There's no such thing as return/parameter types. Its up to you to implement
a mechanism to do this if you want it.

N





From tmk at mayo.edu  Wed Aug 25 15:11:36 1999
From: tmk at mayo.edu (Tom Kinter)
Date: 25 Aug 1999 19:11:36 GMT
Subject: PIL: problem with im.show()
Message-ID: <7q1f58$r7v$1@tribune.mayo.edu>

system configuration
  Sun Solaris-2.6
  python-1.5.2
  PIL-1.0

in PIL/libImaging, the following step suggested in the installation works:
  make check

in PIL/Scripts the following scripts were tried and worked:
  python viewer.py lena.ppm
  python thresholder.py lena.ppm
  python enhancer.py lena.ppm
  python painter.py lena.ppm

the following interactive process 
>>> import Image
>>> im = Image.open("lena.ppm")
>>> print im.format, im.size, im.mode
PPM (128, 128) RGB
>>> im.show()
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "/usr/local/lib/python1.5/site-packages/PIL/Image.py", line 694, in show
    _showxv(self, title, command)
  File "/usr/local/lib/python1.5/site-packages/PIL/Image.py", line 974, in _showxv
    file = self._dump(format=format)
  File "/usr/local/lib/python1.5/site-packages/PIL/Image.py", line 311, in _dump
    file = file + "." + format
TypeError: illegal argument type for built-in operation
>>> 

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  Tom Kinter     tmk at mayo.edu     507-284-4981
     Mayo Foundation Rochester MN 55905 USA
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~





From mal at lemburg.com  Fri Aug 20 03:52:59 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Fri, 20 Aug 1999 09:52:59 +0200
Subject: Viewing contents of an object
References: <01beeaa5$c7f6db60$7f8345c6@TheBlueWizard>
Message-ID: <37BD095B.7100A3CF@lemburg.com>

The Blue Wizard wrote:
> 
> Hi Pythoneers!
> 
> I would like to be able to view the content of an arbitrary object, e.g.
> view(myObject) would print
> class: myClass
> class variables:
>     counter   = 6
>     topicoftheday = "Python is fun!"
> instance variables:
>     x = 4
>     y = 7
> 
> Thanks!

Try hack.py which you can download from my Python Pages. You can
use that code to build your own versions...

>>> hack.show(p,5)
<Proxy.Proxy.InstanceProxy for __main__. ...
    __class__ :
        <class Proxy.Proxy.InstanceProxy at 80ea ...
            __name__ :
                'InstanceProxy'
            __bases__ :
                ()
            __dict__ :
                __doc__ :
                    ' Proxy that wraps Python instances tran ...
                __frozen__ :
                    1
                __getattr__ :
                    <function __getattr__ at 80f7838>
                __init__ :
                    <function __init__ at 80f41e8>
                __module__ :
                    'Proxy.Proxy'
                __repr__ :
                    <function __repr__ at 80f7818>
                __setattr__ :
                    <function __setattr__ at 80eaea8>
    __dict__ :
        proxy_getattr :
            <built-in method proxy_getattr of Proxy  ...
                __name__ :
                    'proxy_getattr'
                __members__ :
                    __doc__ : 'proxy_getattr(name)\012\012Tries to get ...
                    __name__ : 'proxy_getattr'
                    __self__ : <Proxy object at 80ecce8>
        proxy_object_repr :
            '__main__.C'
        proxy_setattr :
            <built-in method proxy_setattr of Proxy  ...
                __name__ :
                    'proxy_setattr'
                __members__ :
                    __doc__ : 'proxy_setattr(name,value)\012\012Tries  ...
                    __name__ : 'proxy_setattr'
                    __self__ : <Proxy object at 80ecce8>

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   133 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/





From tdixon.no at spam.fwi.com  Fri Aug 13 17:45:53 1999
From: tdixon.no at spam.fwi.com (Timothy Dixon)
Date: Fri, 13 Aug 1999 21:45:53 GMT
Subject: Does Python run under Windows 2000? (build 2031)
References: <37B40D58.DD6AADC0@RobotVision.com> <37B437F8.D07EDEB8@zipzap.ch>
Message-ID: <37b491c4.89061393@midpoint>

On Fri, 13 Aug 1999 17:21:28 +0200, Boris Borcic <zorro at zipzap.ch>
wrote:

>James E. Green wrote:
>> 
>> I'm running Windows 2000 RC1 (build 2031, the very latest).
>
>2000 ? We aren't there yet. Apparently your OS suffers from Y2K.
>
>BB

Windows 2000 is in beta.  It's really NT 5, but apparantly they didn't
want the solid reputation of NT to be associated with it so they
adopted the numbering scheme from the buggier set of operating
systems. :-)




From thomsensven at my-deja.com  Thu Aug 12 08:00:48 1999
From: thomsensven at my-deja.com (thomsensven at my-deja.com)
Date: Thu, 12 Aug 1999 12:00:48 GMT
Subject: Problems importing Modules under win32
Message-ID: <7oud1c$vcd$1@nnrp1.deja.com>

Hi,

I wanted to try out saxlib, but I can't import it under Windows NT:

dir-structure:

C:
|
- Users (dir)
  |- thomsen (dir)
     |- xml-python (dir)
        |-saxlib (dir)
          |-Saxlib (dir)
            |-xml (dir)
              |-sax (dir)
                |-saxexts (file)

python-log:

>>> import sys
>>> sys.path
['' [stuff deleted] 'C:\\Users\\thomsen\\xml-python\\saxlib\\Saxlib\\',
'C:\\Users\\thomsen\\xml-python\\saxlib\\Saxdrv', [stuff deleted], 'C:
\\Program Files\\Python']
>>> import xml.sax.saxexts
Traceback (innermost last):
  File "<stdin>", line 1, in ?
ImportError: No module named xml.sax.saxexts

As you can see, the path is correctly included, when I try to import
a testfile from C:\Users\thomsen\xml-python\saxlib\Saxlib it works.
When I put saxexts directly into e.g. ...../Python/Lib it is
found, but that's no solution....

Any suggestions?

Greetings,

Sven



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From news at dorb.com  Mon Aug 16 20:52:54 1999
From: news at dorb.com (Darrell)
Date: Mon, 16 Aug 1999 20:52:54 -0400
Subject: function which knows who has called it.
References: <934841569snz@vision25.demon.co.uk>
Message-ID: <Nn2u3.2080$8P.47694@typhoon.nyroc.rr.com>

You can get the callers name from this also.
But that's up to you.

def myDirExcept(howFarBack):
        """
        Get the directory of the module running at "howfarBack "
        in the stack frames
        This value is compiled into the .pyc file
        This can be a problem.
        Name will change depending on where python was run from
        when the pyc file was created
        Unix can store a relative path
        Windows stores an absolute path
        """
        try:
                raise ZeroDivisionError
        except ZeroDivisionError:
                f = sys.exc_info()[2].tb_frame

        for i in range(howFarBack):
                f = f.f_back

        t = f.f_code
        fname = t.co_filename
        p = string.split(fname, os.sep)
        fname = string.join(p[:-1], os.sep)
        return fname


--
--Darrell
Phil Hunt <philh at vision25.demon.co.uk> wrote in message
news:934841569snz at vision25.demon.co.uk...
>
> Is if possible in Python to have a function that prints the name
> of the function that called it. (I'm thinking of using this in
> some debugging code I want to write).
>
> --
> Phil Hunt....philh at vision25.demon.co.uk
>





From 71351.356 at compuserve.com  Tue Aug 17 21:05:04 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Tue, 17 Aug 1999 21:05:04 -0400
Subject: stat, chmod and RDONLY file (win32)?
References: <7pambk$3m8$1@ssauraaa-i-1.production.compuserve.com> <37B9D5F6.2AC782E1@compaq.com>
Message-ID: <7pd13e$if1$1@ssauraab-i-1.production.compuserve.com>

Thanks thats the info.   I tried a quicky:   os.chmod( name, 0777) and it
works.    Funny I associated the four octal flags with Unix, not win98 or
NT.    Oh well, newbieness....

Greg Ewing <greg.ewing at compaq.com> wrote in message
news:37B9D5F6.2AC782E1 at compaq.com...
> "T. C. Mits" wrote:
> >
> > mean that one must supply the c-run-time bit flags,
> > documented elsewhere (where?), as mentioned in the open() API blurb,
such as
> > O_RDONLY.
>
> No, O_RDONLY and friends are a completely different set
> of modes (confusing, I know!)
>
> The sort of mode you want here consists of 3 octal
> digits, for owner, group and other. Each digit
> contains read, write and execute permission bits
> in that order.
>
> So, to add write permission for the owner, for
> instance, "or" the existing permissions with
> octal 0200.
>
> There's also a fourth octal digit to the
> left which contains the setuid, setgid and
> sticky bits, but these aren't relevant for what
> you're doing here.
>
> Hope that helps,
> Greg





From mnot at mnot.net  Tue Aug 17 18:20:40 1999
From: mnot at mnot.net (Mark Nottingham)
Date: Tue, 17 Aug 1999 22:20:40 GMT
Subject: unsubscribing from the *old* list
Message-ID: <033401bee8fe$bd935840$0301a8c0@mnot.net>

OK, I've tried sending much mail to python-list-request at cwi.nl, with no
results. How do I get off this beast? I remember an old mail explaining how
to transition from the old list to the new, but for the life of me I can't
remember it...

ta,




From janne at nnets.fi  Tue Aug 24 03:54:19 1999
From: janne at nnets.fi (Janne Sinkkonen)
Date: 24 Aug 1999 10:54:19 +0300
Subject: RPM for Python 1.5.2 ?
References: <slrn7qrn3u.7fj.dyoung@baltazar.baltazar.org> <37ADE4C1.194C31D6@lemburg.com> <19990809113731.A6757@baltazar.baltazar.org> <37AFD5A9.4C45332F@lemburg.com> <7ps5t7$2u6$1@nntp.Stanford.EDU>
Message-ID: <m3u2ppd36c.fsf@kinos.nnets.fi>

amitp at Xenon.Stanford.EDU (Amit Patel) writes:

> For normal use I don't use Tkinter so it's not affecting me.  All I
> wanted to do with Tkinter was to try out IDLE.. :) I'm quite happy
> with Oliver's RPMs.. it's much more convenient than asking everyone at
> work to compile Python 1.5.2 on their system. :)

BTW, you need to edit the netcfg script of RH (5.2 or 6.0) to get it
to work with 1.5.2. That's easy though, just adding '=None' to two
function arguments (which are not originally keyword arguments but are
preceded by keyword arfuments).

-- 
Janne Sinkkonen      <janne at iki.fi>      <URL: http://www.iki.fi/~janne/ >



From gtnorton at my-deja.com  Mon Aug 30 23:45:21 1999
From: gtnorton at my-deja.com (gtnorton at my-deja.com)
Date: Tue, 31 Aug 1999 03:45:21 GMT
Subject: Grids in Tkinter
References: <936050813snz@vision25.demon.co.uk>
Message-ID: <7qfj4f$rf7$1@nnrp1.deja.com>

In article <936050813snz at vision25.demon.co.uk>,
  philh at vision25.demon.co.uk wrote:
>
> Can anyone point me to anywhere that documents how Grids work in
> Tkinter? A short example program would be nice.
>
> --
> Phil Hunt....philh at vision25.demon.co.uk
>
>
I'm sure Fredrik Lundh's Introduction to Tkinter has what your looking
for:             www.pythonware.com/library.htm

                        A Great Read!
                                     Good Luck,
                                              gtnorton


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From I.Clarke at strs.co.uk  Fri Aug 20 04:41:02 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Fri, 20 Aug 1999 09:41:02 +0100
Subject: Help!  Dialling on modem
References: <7pemm3$fd5@enews3.newsguy.com>
Message-ID: <37BD149D.23EA9B0D@strs.co.uk>

I had to do this using Java once.  I suggest you generalise your search
to "how to send and receive from the serial ports" - unless you want to
connect to the Internet, in which case you should let Windows worry
about dialling out.

Ian.

Michael Spencer wrote:
> 
> Have looked thru docs & examples on networking. Lots of info on using
> sockets and TCP. But nothing on how how to cause the modem to dial out.
> 
> Lots of good reasons for not using M$ TAPI.
> 
> Can anyone tell me where to find sample code on using modem to dial a
> telephone number under Win32?
> 
> Thanks
> 
> Mike



From skip at mojam.com  Mon Aug 16 01:53:10 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 16 Aug 1999 00:53:10 -0500 (CDT)
Subject: Python CGI Scripts
In-Reply-To: <Pine.LNX.4.04.9908150140180.584-100000@gnr.u2me3.com>
References: <7p5gu7$nlq@dfw-ixnews14.ix.netcom.com>
	<Pine.LNX.4.04.9908150140180.584-100000@gnr.u2me3.com>
Message-ID: <14263.42677.956504.870922@dolphin.mojam.com>

    Stephen> But... have a lot of people adopted python as their new cgi
    Stephen> language? 

Just one data point from one person (who never wrote a single Perl CGI
script, so it's my old cgi language as well):

    ~% ls ~httpd/cgi-bin/* | xargs file | egrep -i python | wc -l
	 51

and I tend to use ZServer (see http://www.zope.org/) much more these days
instead of CGI.  

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From tim_one at email.msn.com  Sun Aug 29 21:13:12 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 29 Aug 1999 21:13:12 -0400
Subject: Porting Python
In-Reply-To: <Qtey3.430$LZ.46228@ndnws01.ne.mediaone.net>
Message-ID: <000101bef284$d4bccb20$a22d153f@tim>

[Dave Abrahams]
> Suppose I wanted to build a version of python for a platform that's
> not yet supported. Does anyone have concise info on where the OS-
> dependencies are?  I'd like to get a sense of the scope of the job.

It's not OS-dependence so much as libc dependence -- Python does very little
under the covers that's not straight ANSI C.  The exceptions are generally in
modules that you don't *need* to build in (like threads and sockets -- and
those are designed to be "configured out" painlessly).

The only pervasive dependence is on malloc, realloc and free.  calloc is
declared for some historic reason, but last I checked is never actually
referenced.

Beyond those, Python needs enough of stdio to do text mode input of source
files and text mode output for "print", & binary I/O of compiled .pyc and .pyo
files.  So it needs too to open files, close them, and tell whether they
already exist.  This makes for obvious problems on platforms without
filesystems.

It assumes sizeof(int) is at least 4, but does not assume sizeof(long) is
bigger, or that sizeof(void*) is either of those, or that integral types and
pointers can be cast to the other (it plays no tricks at all with pointers).

People generally don't report major porting problems unless the platform C
implementation is insane; e.g., someone recently tried to port to an embedded
system whose OS and C runtime cleverly arranged to make all static vrbls shared
across processes(!).  Python doesn't have much static data, but what it does
have, it recklessly <ahem> assumes it owns.

What else?  Python likes to use memory, but you can feed it eat either live
rats or frozen ones without undue complaint.

pretending-we-never-met-ly y'rs  - tim





From gmcm at hypernet.com  Sat Aug 28 17:31:04 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sat, 28 Aug 1999 16:31:04 -0500
Subject: IDLE and iso-8859-2; threading tutorial
In-Reply-To: <slrn7sgenr.n2.smithy@hoohoo.org>
Message-ID: <1276250498-42824288@hypernet.com>

?ukasz Kowalczyk writes:

> Also, is there a tutorial about using threads in Python? I know
> there are a few examples in doc/Demos directory but I would like to
> read about real-world problems that may arise. I still remember how
> much time I spent looking for bugs in my Java threading program and
> finally found out that putting Thread.yield() all over the code did
> the trick...

You must've been using a platform where Java didn't use real threads 
(ie, it used user-level threads instead of system-level threads). 
Python doesn't attempt to fake threads. It uses the system's threads, 
or none at all.

So if you use the threading module, you already know how to use 
Python threads - they're modeled after Java threads (except you never 
have to yield()).

- Gordon



From olevi at my-deja.com  Tue Aug 17 03:02:38 1999
From: olevi at my-deja.com (Ole-Kristian Villaboe)
Date: Tue, 17 Aug 1999 07:02:38 GMT
Subject: Unsorting(randomizing) a sequence
Message-ID: <7pb1eb$elt$1@nnrp1.deja.com>

Is there a command to do this? (Can't find one but...)
I've been writing an algorithm myself, but it's incredibly slow and I
would like something quicker (of course :) )

Ole


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From winkjj at my-deja.com  Fri Aug 13 16:06:50 1999
From: winkjj at my-deja.com (winkjj at my-deja.com)
Date: Fri, 13 Aug 1999 20:06:50 GMT
Subject: Python API for TET
Message-ID: <7p1tsi$dqe$1@nnrp1.deja.com>

Is anyone working on or has anyone heard of someone working on a Python
API for TET (Test Environment Toolkit)?


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From graham at sloth.math.uga.edu  Sun Aug 15 17:09:39 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 15 Aug 1999 21:09:39 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au>
Message-ID: <7p7aaj$sdl$1@cronkite.cc.uga.edu>

John (Max) Skaller (skaller at maxtal.com.au) wrote:
: 	Preliminary tests indicate reasonable results
: determining types. In critical cases such as inner loops,
: it is usually possible:
: 
: 	s = ""
: 	for  i in range(1,10): s = s + str(i)
: 
: Obviously i is an integer here, and s a string.

Is it? Surely that depends on what "range" and "+" are bound to at
runtime.

I don't wish to be a spoil-sport but I really do believe that type
inference in Python is a waste of time, and hence your ideas to speed
things up are doomed to fail in all but the simplest of cases. Python
is far too dynamic to be type inferenced (or even type checked). Whether
that's a good thing or not is another matter, but it's certainly a
fact I think.

graham

-- 

      England's not the mythical land of madame george and roses
        It's the home of police who kill black boys on mopeds



From skaller at maxtal.com.au  Sun Aug 15 12:24:24 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sun, 15 Aug 1999 16:24:24 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <xj0vhamuzwh.fsf@cyclone.harvard.edu>
Message-ID: <37b6e7cf.345281819@news.triode.net.au>

On 11 Aug 1999 10:59:10 -0400, Michael Haggerty <mhagger at blizzard.harvard.edu> wrote:

>skaller at maxtal.com.au (John (Max) Skaller) writes:
>> 	3) 'in' applies to dictionaries 'as if a sequence of keys were used'
>
>At first I thought `hey that's cool!' since I'm kindof sick of typing
>has_key() all the time.  But then I realized that this is not really
>consistent with sequence types.
>
>    item in list     IMPLIES      list[something] == item
>
>but according to your definition,
>
>    item in dict     IMPLIES      list[item] == something
>
>where in each case `something' is undetermined.  I suggest you rethink
>this `feature'.

	I agree there are some difficulties here, which is why
I'm appealing to the Python community for ideas.

	There is an alternate possible definition:

	k,v in dict

which might make more sense:

	dict.has_key(k) and dict.get(k)

However the idea is that 

	x in Y

means 'x is a member of the set of elements of Y'
and for a list the 'set of elements of Y' is the
set of elements of the list, and for a dictionary
it is the set of keys of the dictionary.

In this way, the meaning is consistent for lists
and dictionaries. Furthermore,

	for x in Y: S(x)

is also well defined as executing S(i) for
each i in Y: for lists and tuples, in the order
of the list or tuple, for dictionaries either
in a random order, or in order of the sorted
keys.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From amaranda at spider.com  Wed Aug 25 09:29:17 1999
From: amaranda at spider.com (Alex Maranda)
Date: Wed, 25 Aug 1999 14:29:17 +0100
Subject: Making A Standalone Executable Under Windows NT/95/98
References: <hUQw3.117$Kl.24641667@speed.city-net.com>
Message-ID: <37C3EFAD.7A7@spider.com>

jhomme.. wrote:
> 
> Is this possible?  If so, How?
I never tried it but it should doable. You'll have to use 'freeze' of
course, but the compliling and Makefile part's going to be tricky. And
it depends on which compiler you're using.

If you have VC++:
Python on Windoze is built with VC++. You'll need to get freeze from the
source package. Provided that (big if) freeze generates a nmake
compatible makefile, you'll only need to have the env vars for VC++ set
(I remember there is a .BAT named vcvars32 which appends them to
AUTOEXEC.BAT). Of course the CygWin user tools (bash, ls...) are highly
recommended, unless you want to put up with the DOS shell. After you
freeze your app the usual way, I guess it's just a matter of invoking
nmake on the generated makefile. If freeze doesn't know the nmake format
and it generates a Unix-style makefile, then good luck, you'll have to
convert it manually.

If you have CygWin/Mingw32:
Well, you have the whole GNU tool-chain, more or less; the process
should be similar with what you do on Unix. Except that the .OBJ formats
are not compatible between VC++ and GNU, so you can't use the stock lib
from the binary distro of Python, you'll have to rebuild it from source.
I don't know if this is supported.

If you figure it out, please let me know, I'll want to do it myself in
the near future (cosmic grief).

Cheers,
Alex
-- 
Alex Maranda         mailto: amaranda at spider dot com
Spider Software Ltd. Tel: +44 (0)131 4757036
Edinburgh, UK        http://members.xoom.com/Alex_Maranda
STREAMS based communications protocols for embedded systems



From Christian.Lemer at usrconsult.be  Fri Aug  6 10:51:17 1999
From: Christian.Lemer at usrconsult.be (Christian Lemer)
Date: Fri, 06 Aug 1999 16:51:17 +0200
Subject: Python win32com/Excel newbie question
Message-ID: <37AAF665.CB6DB912@usrconsult.be>

I currently use python to generate reports in Excel. It works find but I
can't figure out how to set up some of the 'Variants' that seems to be
able to receive a numerical value AND a boolean value

i.e.: xlSheet.PageSetup.Zoom

It seems that in VisualBasic you may say:

	xlSheet.PageSetup.Zoom = True
	xlSheet.PageSetup.Zoom = 300

In fact, the Zoom feature has an edit box but is also part of a
radiobutton group. It seems that setting to True/False changes the
radiobutton state and setting the numerical value changes the value
of the edit box.

In Python, I can assing the numerical value but can't figure out how
to set the True/False state.

Thanks in advance for your help.

-- 
Christian LEMER                                    tel +32.10.65.44.11
UsrConsult S.P.R.L                                 fax +32.10.65.44.10
rue Margot, 37	                              http://www.usrconsult.be
B-1457 Nil St Vincent             mailto:Christian.Lemer at usrconsult.be



From arcege at shore.net  Tue Aug 17 11:35:39 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 17 Aug 1999 15:35:39 GMT
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org> <14265.5096.537800.149227@buffalo.fnal.gov> <37B91E9A.2451936C@w3.org> <001301bee891$82fccc60$f29b12c2@secret.pythonware.com> <1Feu3.381$mV.141087@news.shore.net> <14265.31990.925778.245319@dolphin.mojam.com>
Message-ID: <fjfu3.386$mV.141087@news.shore.net>

Skip Montanaro <skip at mojam.com> wrote:

: Michael> The better question is:
: Michael>    Why did it change from documenting these components in the OS-
: Michael>    dependant sections (specifically only posix) to documenting them
: Michael>    mostly in the module-os section in the current release?

: I think it was because they were trying to steer people toward

:     import os

: so new users didn't know to look in the posix chapter for the docs.

That was my point. ;)  The "powers-that-be" think the idiom that should
be used is "import os".  It does not mean that "import posix" cannot be
used, but that you should know what you are "giving up" if you do.

  -Arcege




From joeb at go2net.com  Fri Aug  6 23:53:21 1999
From: joeb at go2net.com (Joe Bowbeer)
Date: Fri, 06 Aug 1999 20:53:21 -0700
Subject: lambda fun: infinite streams
Message-ID: <37ABADB1.834A664E@go2net.com>

Fya.

I was doing a little investigating with Python's lambda expressions and
it looks like they are indeed powerful enough to create some of my
favorite beasts: infinitely long streams [*].

If we define a stream to be a 'pair' of the form (head, tail), where
head is the first element and tail is a function that returns the rest
of the stream (which is itself a stream), then we can define 'head' and
'tail' functions in Python as follows:

    def head(s): # return first element in stream
        return s and s[0]

    def tail(s): # return remainder of stream
        return s and s[1] and s[1]()

We create a new stream given a head element and a tail stream as
follows:

    s = (head, lambda : tail)

    # I'd use a 'special form' or macro above if Python had them..

Here are two more functions that operate on streams. (I'd implement
these tail-recursively if it were more efficient.)

    def nthStream(s, n): # return nth element of stream s
        while n > 0:
            s, n = tail(s), n-1
        return head(s)

    def printStream(s, n): # print first n elements of stream s
        while n > 0:
            print head(s),
            s, n = tail(s), n-1
        print

Now things start to get interesting:

    # infinite stream of ones
    ones = (1, lambda : ones)

    # print some ones
    printStream(ones, 10)

    def addStreams(si, sj): # add corres. elements of two streams
        if not si: return sj
        if not sj: return si
        tailsum = lambda ti=si, tj=sj : addStreams(tail(ti), tail(tj))
        return (head(si) + head(sj), tailsum)

    # infinitely many integers
    integers = (0, lambda : addStreams(ones, integers))

    # infinite stream of Fibonacci numbers
    fibs = (0, lambda : (1, lambda : addStreams(tail(fibs), fibs)))

    # Too bad 'int' can only represent a finite number of digits...

Better performance can be achieved by memo-izing the functions that
evaluate the tails. A simple Memo class can be defined as follows. It
lazily evaluates the 'thunk' function and then memorizes the result:

    class Memo: # memoized call-by-need thunk
        def __init__(self, thunk):
            self.result, self.thunk = None, thunk
        def eval(self):
            if self.thunk:
                self.result, self.thunk = self.thunk(), None
            return self.result

Memo-ized streams are created as follows:

    s = (head, Memo(lambda : tail))

And the tail function becomes:

    def tail(s): # return remainder of stream
        return s and s[1] and s[1].eval()

----
*SICP:

    Structure and Interpretation of Computer Programs
    Abelson and Sussman
    http://mitpress.mit.edu/sicp/

----
Joe Bowbeer






From milese at pacbell.net  Sat Aug 14 03:08:47 1999
From: milese at pacbell.net (Miles Egan)
Date: Sat, 14 Aug 1999 00:08:47 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <37B50C9D.868F0F02@inka.de>
Message-ID: <37B515FF.68DBD6F4@pacbell.net>

> Of course YMMV. This is up to you. But I would suggest learning at least
> one FP language is one of the best things a programmer can do. Possibly
> there are other suprises around maybe Prolog would be another good idea?

This is excellent advice.  Even if you never write a line of production
code in Scheme or ML, studying these languages can be very
enlightening.  Python and Perl are more alike than different in the
grander family of programming languages.  I particularly recommend "The
Structure and Interpretation of Computer Programs" by Abelson and
Sussman.  I'm still blown away by the simplicity of their implementation
of a Scheme evaluator.

miles



From robin at jessikat.demon.co.uk  Thu Aug 26 21:40:35 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Fri, 27 Aug 1999 02:40:35 +0100
Subject: Library philosophy
References: <37C57A80.D5C62704@prescod.net>
 <14277.39227.719967.140778@dolphin.mojam.com>
Message-ID: <DjSxbBATyex3EwOE@jessikat.demon.co.uk>

In article <14277.39227.719967.140778 at dolphin.mojam.com>, Skip Montanaro
<skip at mojam.com> writes
>
...
>This is a known issue.  I'll put words in g--d-'s mouth (and let him take
>them out when he returns) once again and indicate that the thin libraries
>were made that way on purpose.  I can't pretend to understand all the
>reasons that is the case, but it appears to be a pretty consistent approach.
>It also makes the most sense to me when you are exposing a C API with an
>extension module.  If you come up with some higher-level classes (e.g. a
interestingly tcl took exactly the opposite view recently with sensible
stuff being  deferred or reduced to common denominators. The sockets
stuff was brain dead on windows etc etc. The commands looked the same,
but often there were hidden semantic/operational differences. Tk on the
other hand has been moving the opposite way. Instead of one look for all
platforms we get more and more 'native' behaviour.
>logical filesystem that doesn't expose chmod and friends and that acts
>uniformly across different OS platforms), my guess is that it would be
>looked upon favorably.
>
>    Paul> Would it it make sense to start wrapping the low-level modules
>    Paul> with high-level, easier to use modules that would be "the default"
>    Paul> in Python 2?
>
>Makes good sense to me.
>
me too
-- 
Robin Becker



From mal at lemburg.com  Mon Aug 16 09:15:32 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 16 Aug 1999 15:15:32 +0200
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <37b5a8e0@cs.colorado.edu> <7p7d22$1hh$1@holly.csv.warwick.ac.uk> <004f01bee7bc$f9945b70$f29b12c2@secret.pythonware.com>
Message-ID: <37B80EF4.7F571AD4@lemburg.com>

Fredrik Lundh wrote:
> 
> Andrew Clover <esuzm at primrose.csv.warwick.ac.uk> wrote:
> > I'd like a syntax like:
> >
> >   x.+(1)
> >
> > With the idea that x is an object (of type int) with a '+' method that
> > adds stuff to it. It'd be nicer without the brackets though.
> 
> mutable integers?  now that's a really cool feature! (known
> from Fortran and Tcl, among others. I'm pretty sure Perl
> has them too ;-).

Here's a mutable integer implementation: it's a counter
type with which you can do cute things like:

>>> c = Counter(0)
>>> ++c
1
>>> ++c
2

You can even split the increment into two operations (well, that's
due to the hack used to implement this):

>>> +c
2
>>> +c
3

Here's some more magic:

>>> +-c
3
>>> -+c
4
>>> -+c
4
>>> -+c
4
>>> +-c
3
>>> +-c
3

To find out how this works, see:

	http://starship.skyport.net/~lemburg/mxCounter-0.1.zip

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   138 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From bwarsaw at cnri.reston.va.us  Thu Aug 19 15:31:17 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Thu, 19 Aug 1999 15:31:17 -0400 (EDT)
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com>
	<37BC75F0.B482DA6D@starmedia.net>
	<37bc4352@cs.colorado.edu>
Message-ID: <14268.23429.917082.290102@anthem.cnri.reston.va.us>

>>>>> "TC" == Tom Christiansen <tchrist at mox.perl.com> writes:

    TC> The fact that they did icky windows help and not man pages
    TC> says something significant.

It says that Windows users were more motivated to provide platform
specific documentation than any Unix user so far.

-Barry



From mwh21 at cam.ac.uk  Fri Aug 27 13:42:42 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 27 Aug 1999 18:42:42 +0100
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan> <slrn7s408d.gvn.wtanksle@dolphin.openprojects.net> <hhvha5tswc.fsf@forcix.roof.lan> <14274.54686.281949.239343@dolphin.mojam.com> <7q0jdl$k53$1@nnrp1.deja.com>
Message-ID: <m3aerd15nx.fsf@atrus.jesus.cam.ac.uk>

grahamclemo at my-deja.com writes:

> After a bit of experimentation I found you can replace except: with
> except 0: (I doubt anyone has raise 0 in their code) and except
> ZeroDivisionError: with except 0:#ZeroDivisionError: etc.

If you did have "raise 0" in your code, chances are it wasn't doing
what you were expecting:

>>> try:
...  raise 0
... except 0:
...  print "caught"
...
Traceback (innermost last):
  File "<stdin>", line 2, in ?
TypeError: exceptions must be strings, classes, or instances

so "except 0:" is s good block-that-will-never-catch. One for the back
of the mind, I think.

> (I'm surprised Python silently accepts *any* value in except
> statements)

It's probably too awkward (& not particularly helpful) to ban them.

Regards,
Michael



From jeremy at cnri.reston.va.us  Thu Aug 19 12:05:52 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Thu, 19 Aug 1999 12:05:52 -0400 (EDT)
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <wkyaf7kgsh.fsf@turangalila.harmonixmusic.com>
References: <7pb1eb$elt$1@nnrp1.deja.com>
	<14265.29899.470021.309355@dolphin.mojam.com>
	<y0jiu6edoa9.fsf@vier.idi.ntnu.no>
	<xj0d7wl9acd.fsf@cyclone.harvard.edu>
	<37BAB643.9D4E861E@strs.co.uk>
	<wkyaf7kgsh.fsf@turangalila.harmonixmusic.com>
Message-ID: <14268.11104.358158.638005@bitdiddle.cnri.reston.va.us>

>>>>> "DS" == Dan Schmidt <dfan at harmonixmusic.com> writes:

  DS> Ian Clarke <I.Clarke at strs.co.uk> writes:

  | I did some timing tests using the code I have attached below, here are
  | the results (in seconds) for sorting a list of 200 items 10 times:
  | 
  | Dan Schmidt's solution (using sort): 14.98
  |
  | def s1(list):
  |     """Dan Schmidt's solution - (mis)using sort"""
  |     def randomize(a,b):
  |         return random.choice([-1, 0, 1])
  |     list.sort(randomize)
  |     return 

  DS> That was Jeremy Hylton's solution.  I just complained about it!

Let's not attach anyone's name to it :-).  I like it because it's a
obscure/silly approach, not because it produces a good result.  It
sounds like a shorter way to spell it, though, would be:
    list.sort(lambda x,y:random.choice([0,1]))

Jeremy





From neelk at brick.cswv.com  Thu Aug 19 20:22:16 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 20 Aug 1999 00:22:16 GMT
Subject: 'if' as a _function_, in Scheme and Python
References: <7pi0kl$ko8$1@nnrp1.deja.com>
Message-ID: <slrn7rpaio.7pl.neelk@brick.cswv.com>

On Thu, 19 Aug 1999 22:31:49 GMT, oleg at pobox.com <oleg at pobox.com> wrote:
>In most languages, 'if' has a distinguished status, of a statement or
>a special syntax form. One wonders whether 'if' can be a regular
>function. 
>
>For example, in Scheme:
>        (define (my-if condition then-branch else-branch)
>           ((or (and condition then-branch) else-branch)))
>
>on stipulation that the arguments of this function are wrapped in
>lambda-expressions (i.e., thunks):

You should look also look at Smalltalk; all operations are implemented
with message sends, and conditionals are implemented exactly as you
describe (ie with closures). Python is more like a Smalltalk than a
Scheme (to the extent that's meaningful -- ST and Scheme are
practically duals in language space). 

>x = eval((lambda: exp1, lambda: exp2)[not condition].func_code,vars())
>
>this expression is equivalent to
>        if condition: x = exp1
>        else: x = exp2

Wow. That's probably the best ?: emulation yet. "Best" being defined
as "makes me laugh with appreciation," as opposed to actually being
practical; the if-else construction is the only one that J. Random
Hacker is likely to understand when he comes across in actual code.


Neel



From fredrik at pythonware.com  Fri Aug 20 08:36:46 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 12:36:46 GMT
Subject: making python man pages looks hard
References: <1103_935134343@shoal>
Message-ID: <006801beeb08$aafe2280$f29b12c2@secret.pythonware.com>

(posted and mailed)

Dan Connolly <connolly at w3.org> wrote:
> I happen to be a *big fan* of the python documentation[2] as is
> but I don't see any reason why it shouldn't be available as
> man pages too. Or at least... I didn't.

> The source format is LaTeX. Converting from LaTeX is hard/messy:
> 
> ...
> For example, I conjecture that it is impossible to write a program that
> will extract the third word from a TeX document. It would be an easy
> task for 80% of the TeX documents out there -- just skip over some
> formatting stuff and grab the third bunch of characters surrounded by
> whitespace.
> ...

luckily, Fred Drake has already made most of the hard work
here -- check the Doc/tools/sgmlconv directory:

...

$ more Doc/tools/sgmlconv/README
These scripts and Makefile fragment are used to convert the Python
documentation in LaTeX format to SGML.  XML is also supported as a
target, but is unlikely to be used.

This material is preliminary and incomplete.

...

not sure how incomplete, though.  I don't have things setup
so I can try the current release of this (from the CVS archive),
but maybe Fred can give us a status update?

(maybe he has, I cannot say I've bothered to read *all* the
messages on the group today...  sigh...)

anyway, if this stuff works, writing an esis2man converter
cannot be *that* hard.

</F>




From xah at best.com  Sun Aug 22 07:51:08 1999
From: xah at best.com (Xah)
Date: Sun, 22 Aug 1999 04:51:08 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
	 <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com>
Message-ID: <37bfe42c$0$221@nntp1.ba.best.com>

>From: miker3 at ix.netcom.com (Michael Rubenstein)
>...
> 
> The problem in comp.lang.perl.misc is not the beginners -- it is
> the regulars.

We, as Perl programers, enjoy what we do. Flaming is part of the culture,
not to mention a good training ground for stupid newbies. Rather a nice
multi-purpose filtering system. If you have thick skin and you endured, then
you have what to takes to be a Perl preacher like us. Otherwise, we're more
than happy to kick you back to your dustless hospital room or research labs.

Perl wouldn't be a nice community without bad asses like Tom Christensan,
Chris Nador and some other regulars I don't want to mention now. We thrive
on smiting newbies. We live to smite each other too. If you start to feel
uncomfy,  get out.

un'stand?

PS I love Tommy's automatic email manifesto on newsgroup line formats. I got
few of them in the past, and I lived to tell the tale. One day I'll write a
bot that bounce one for one of every his spams. Here in comp.lang.perl.misc,
we learn by doing.

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 "I -- like many unix weenies -- love stupid nonsensical quotes."
       --autononymous



From garethr at cre.canon.co.uk  Mon Aug 23 17:33:32 1999
From: garethr at cre.canon.co.uk (Gareth Rees)
Date: Mon, 23 Aug 1999 21:33:32 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu> <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net>
Message-ID: <siemguw5ar.fsf@cre.canon.co.uk>

xah at weborder.com wrote:
> The ivy tower class of buttheads insist on symbolism for math,
> clinging to the fancy that certain ideas are better expressed in
> certain language.  They created a whole barrier of entry [...]
> 
> [...] If someone cannot read my code, it is their shortcoming.

Can you spot the contradiction?

-- 
Gareth Rees



From I.Clarke at strs.co.uk  Mon Aug  9 08:42:06 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Mon, 09 Aug 1999 13:42:06 +0100
Subject: lambda fun: infinite streams
References: <37ABADB1.834A664E@go2net.com> <slrn7qnsod.jjr.thantos@brimstone.mecha>
Message-ID: <37AECC9E.C0002036@NOSPAM.strs.co.uk>

> Its good to see other Scheme/Fnctional-language freaks hanging around
> with the Python folks, especially those with copies of SICP.

Yeah, Python originally caught my eye because it operated a little like
ML...

Ian.



From g.gruber at xsoft.co.at  Thu Aug 26 06:27:39 1999
From: g.gruber at xsoft.co.at (Gerhard W. Gruber)
Date: Thu, 26 Aug 1999 12:27:39 +0200
Subject: binary data
Message-ID: <37C5169B.824BF1F3@xsoft.co.at>

How do I write binary data to a file?
I'm using open() and write and I always get the error that n is an
integer.

-- 
bye,
   Gerhard
email: sparhawk at eunet.at
       g.gruber at xsoft.co.at



From amaranda at spider.com  Tue Aug 17 05:20:15 1999
From: amaranda at spider.com (Alex Maranda)
Date: Tue, 17 Aug 1999 10:20:15 +0100
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <000101bee869$9058c760$6e2d2399@tim>
Message-ID: <37B9294F.33AA@spider.com>

Tim Peters wrote:
> While I'm tickled by the image, nothing is like SNOBOL4 -- not even its
> modern successor, Icon.  SNOBOL4 was unique, and is still the Best Language
> on Earth for writing string pattern-matching algorithms.
> > ...
> > Did-anyone-we-know-write-a-SNOBOL4-mode-for-the-IBM-026-keypunch?-ly
> > yours,
> 
> yes-but-it-was-in-fortran2-and-kept-blowing-the-tubes-ly y'rs  - tim
Tim, you are the longest running bot in history :-) I hope one day we
the mere humans will get a chance at reading your source code.

some-bots-are-better-than-others-ly y'rs - Alex



From kernr at mail.ncifcrf.gov  Fri Aug 20 20:01:30 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Sat, 21 Aug 1999 00:01:30 GMT
Subject: Thanks for the help. Still left me with a lot of questions but I think I have better direction. What is a troll by the way?
References: <zelv3.13444$x04.816752@typ11.nn.bcandid.com>
Message-ID: <37bdeb1b.31915768@news.erols.com>

On Fri, 20 Aug 1999 19:02:56 -0400, "Daryl Zero"
<Wobbler at greenvillenc.com> wrote:

>Thanks for the help. Still left me with a lot of questions but I think I
>have better direction. What is a troll by the way?

_Troll_ is a USENET term for someone that deliberately posts an
inflammatory message simply to stir up argument.

Don't worry, you aren't one.  Some of us are just a little touchy
because you posted just after a recent flamew^H^H^H^H^H^Hheated
discussion contrasting Python and Perl.

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From nate.pemberton at lmco.com  Thu Aug 26 21:47:48 1999
From: nate.pemberton at lmco.com (Nathan K. Pemberton)
Date: Thu, 26 Aug 1999 18:47:48 -0700
Subject: posixpath normpath with duplicate leading /
Message-ID: <37C5EDD3.6FCA0BD5@lmco.com>

I am running Python 1.5.2 on HP-UX 9.0. The posixpath function
'normpath' seems to handle multiple leading slashes incorrectly. This is
cosmetic, not a functional problem since the file system and/or shell
collapses multiple slashes anyway.

The issue: multiple slashes should always collapse to a single slash,
even as the leading component. This is true on Unix; I'm not sure
whether this would impact DOS.

e.g. how it currently works:
>>> os.path.normpath('////users/nathanp//tmp/.')
'////users/nathanp/tmp'

how it should work:
'/users/nathanp/tmp'

Here is a proposed implementation which seems to work:


# Normalize a path, e.g. A//B, A/./B and A/foo/../B all become A/B.
# It should be understood that this may change the meaning of the path
# if it contains symbolic links!

def normpath(path):
 """Normalize path, eliminating double slashes, etc."""
 import string
 # Treat initial slashes specially
 slashes = ''
 if path[:1] == '/':
  slashes = '/'
 comps = string.splitfields(path, '/')
 i = 0
 while i < len(comps):
  if comps[i] == '.':
   del comps[i]
   while i < len(comps) and comps[i] == '':
    del comps[i]
  elif comps[i] == '..' and i > 0 and comps[i-1] not in ('', '..'):
   del comps[i-1:i+1]
   i = i-1
  elif comps[i] == '':
   if i == 0 or ( i > 0 and comps[i-1] <> ''):
    del comps[i]
  else:
   i = i+1
 # If the path is now empty, substitute '.'
 if not comps and not slashes:
  comps.append('.')
 return slashes + string.joinfields(comps, '/')






From skip at mojam.com  Thu Aug 19 13:50:11 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 19 Aug 1999 12:50:11 -0500 (CDT)
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc313b@cs.colorado.edu>
References: <37bc1339@cs.colorado.edu>
	<19990819174800.D5227@xs4all.nl>
	<37bc313b@cs.colorado.edu>
Message-ID: <14268.16208.414978.281563@dolphin.mojam.com>

    : There are 'raw' strings though, 'r"<string>"', that dont do
    : backslashes.

    Tom> Those aren't the book, and there are no manpages.  Therefore, in
    Tom> many ways, they don't really count.  I know it hurts to hear this,
    Tom> and you're all going to jump on me, but please please think about
    Tom> it for a bit before you do so.

Tom,

You assume the only two proper places to put documentation are in man pages
or in Mark Lutz's book.  Mark's a fine person and wrote a great book, but it
was written when Python 1.4 was the current version.  Raw strings are new
with 1.5.  Was the Camel book always in sync with Perl?  If it was, then
more power to ya.  Perhaps you can share your synchronization secrets with
Mark and Guido.  The Camel book I have documents Perl 4 (so I guess Perl 5
doesn't count).

As for man pages, they aren't the only valid way to document reference
material.  Most Windows and Mac people wouldn't even know what you were
talking about.  From what I can tell by traffic on c.l.p, the bulk of the
Python user base uses Windows these days anyway, so man pages are probably
not a good first choice of documentation format, being largely unavailable
to the unwashed masses.  HTML seems to be pretty ubiquitous these days,
however.

As others have pointed out, the library and language reference manuals (and
the tutorial and a fairly good collection of demo scripts) are available as
part of Python's distribution.  If you have a web browser, a dvi viewer or
PostScript previewer and the appropriate tools to build them, you can browse
them.  If you haven't got the build tools (unlike Perl Python requires a few
non-Python tools - like TeX and latex2html - to build the docs), you can
view them online or download pre-built copies from the Python website.

Is everything in Python's documentation space as complete as the equivalent
Perl docs?  No, and for the forseeable future that will probably be the
case.  I suppose that might make Python less attractive in some people's
eyes.  I view it simply that Python isn't as far along the path to
completion as Perl.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From shecter at darmstadt.gmd.de  Mon Aug  9 09:36:53 1999
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Mon, 09 Aug 1999 15:36:53 +0200
Subject: Q: When and why to use dynamic variables?
Message-ID: <37AED975.2DDF2410@darmstadt.gmd.de>

Hi,

I'm looking for more information about the ability in Python to add
variables to an object instance at runtime.  I'm interested in both
the theoretical as well as practical implications of this feature.

Can anyone point me to information about it?  For example, code where
this feature is used, descriptions of it, suggested uses, etc.

Thanks,
Robb




From ngps at madcap.dyn.ml.org  Fri Aug 20 11:09:24 1999
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 20 Aug 99 15:09:24 GMT
Subject: [ANNOUNCE] M2Crypto 0.01
Message-ID: <37bd6fa4.0@news.cyberway.com.sg>

Hello,

I am pleased to announce the release of M2Crypto 0.01;
the package is available at http://www.post1.com/home/ngps

Following is the README:

 The M2 Python Crypto Toolkit 0.01
-----------------------------------

The M2 Python Crypto Toolkit (M2Crypto) ==
	Python		www.python.org
+	OpenSSL		www.openssl.org
+	SWIG		www.swig.org

Python is an interpreted, dynamic-typing, object-oriented programming 
language that is often compared to Perl, Scheme, Tcl or Java. SWIG -
Simplified Wrapper and Interface Generator - provides glue to link C/C++
libraries with the above-mentioned languages.

M2Crypto presents a Python interface to OpenSSL, via SWIG.

M2 stands for "me, too!"*

This release includes the following: DH, RSA, DSA, RC4, MD5, SHA1,
RIPEMD160, and the EVP interfaces for message digests, HMACs, and 
symmetric ciphers. It requires Python 1.5.2, OpenSSL 0.9.4, and,
optionally, SWIG 1.1p5 or later. It has been tested under FreeBSD 
2.x, Redhat Linux 5.2, and WinNT4. It should run anywhere Python 
and OpenSSL run.

M2Crypto is released under a license similar to Python's. See LICENSE
for details.


For now, M2Crypto is intended to be a prototyping tool; it is _not_ 
for writing production crypto software.

In particular, note the following caveats:

1. There will be memory leaks, because of pointer ownership contention
between Python and SWIG. (These would be due to my misprogramming.)

2. There is no memory locking/clearing for keys, passphrases, etc.

3. The PRNG has no interface for CS**; it is unlikely to be CS***.

4. AFAIK, Python and OpenSSL have not been subjected to the full attention 
of the Bugtraq crowd. M2Crypto's handling of active hostile input is 
probably suspect.

Of course, I hope to address these in future.

Meanwhile, have fun! Your feedback is welcome.


Ng Pheng Siong
ngps at post1.com


* Similar software are Marc-Andre Lemburg's mxCrypto, and two (?)
Python interfaces to the SSL portion of SSLeay/OpenSSL.

** Continuous seeding.

*** Cryptographically strong.

-- 
Ng Pheng Siong <ngps at post1.com>
-- 
Ng Pheng Siong <ngps at post1.com>




From mcfletch at vrtelecom.com  Sun Aug 15 20:29:07 1999
From: mcfletch at vrtelecom.com (Mike Fletcher)
Date: Mon, 16 Aug 1999 00:29:07 GMT
Subject: C's isprint() concept?
In-Reply-To: <Pine.GSO.4.10.9908151727460.3761-100000@crusoe.crusoe.net>
References: <Pine.GSO.4.10.9908151727460.3761-100000@crusoe.crusoe.net>
Message-ID: <NDBBIAPHEKGOBIHCBLIIEEEGCAAA.mcfletch@vrtelecom.com>

If I'm not mistaken, this would be the python equivalent...

teststring = '\000\001that\003\004'

	import re
	NOTPRINTABLE = r'[^ -~]+'
	re.sub(  NOTPRINTABLE, ' ', teststring )
	# replace all sequences of characters
	# not in the character range NOTPRINTABLE

or (from the original form):

	# note use of octal instead hexadecimal escape codes
	# also note use of raw string to prevent null bytes
	# in the regex pattern.
	NONPRINTABLE = r'[\000-\037\200-\377]+'
	re.sub(  NONPRINTABLE, ' ', teststring )
	# replace all sequences of characters
	# in the range NONPRINTABLE

If you were wanting to replace each character with a " ", leave off the plus
character in the range (thereby matching every single character).  If you
need something faster for character-by-character replace, check out the
string module's mapping functions.

I hope that helps,
Mike

-----Original Message-----
From: python-list-request at cwi.nl [mailto:python-list-request at cwi.nl]On
Behalf Of Jeff Pinyan
Sent: August 15, 1999 5:29 PM
To: python-list at cwi.nl
Subject: Re: C's isprint() concept?


>   $string =~ tr/\x00-\x1f\x80-\xff//d;

Perhaps, more readably:

  $string =~ tr/ -~//cd;

  the /c means complement (take the opposite of) the list
  the /d means delete those characters without a replacement
  the list is ' ' through '~', which is the class of printables




From tim_one at email.msn.com  Fri Aug 20 02:47:23 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 20 Aug 1999 02:47:23 -0400
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer)
In-Reply-To: <37bc4daf@cs.colorado.edu>
Message-ID: <001401beead7$db473de0$712d2399@tim>

[posted & mailed]

[tchrist]
> ...
> Python deems def an executable statement.  Perl doesn't.  Therefore,
> it's a gotcha for a Perl programmer, who doesn't understand why the
> silly compiler did that.

That cuts both ways, of course, as does any difference.  That is, I once got
miffed at Perl because my named, nested functions weren't working.  "man
pages" or not <wink>, it wasn't until I studied the source that I figured
out Perl effectively moves all named functions to the top level, and if you
have two nested functions in a package that happen to have the same name,
the first gets tossed into the bit bucket:

$x = 5;

sub outer1 {
    my $x = 50;
    sub inner {
        print "in the first inner, and here's x: $x\n";
    }
    &inner();
}

sub outer2 {
    my $x = 500;
    sub inner {
        print "in the second inner, and here's x: $x\n";
    }
    &inner();
}

&outer1();

Which prints:

    in the second inner, and here's x:

(the first inner vanished, and $x in the second inner is undefined).

It's curious that Python & Perl both *allow* lexical nesting of named
functions, and both do wildly-- yet different! --unexpected things with
them.  Python may not have lexical closures in Perl's sense, but Perl
doesn't have nesting in Python's sense <wink>.

> Apparently, it's just at run-time interring a function that was
> compiled at compile-time in the current symbol table.

Yes, "def" acts exactly like an assignment statement, binding the function
name to the code object in the current local namespace (which happens to be
the module namespace if the function is at top-level, the current function's
local namespace if the "def" is contained in a function, or the class's
namespace if the function is contained in a class).

One cute consequence is stuff like:

if __debug__:
    def f():
        # slow but careful implementation
else:
    def f():
        # fast but caressless implementation

more-powerful-than-a-speeding-#ifdef-ly y'rs  - tim





From mhagger at blizzard.harvard.edu  Tue Aug 17 20:54:58 1999
From: mhagger at blizzard.harvard.edu (Michael Haggerty)
Date: 17 Aug 1999 20:54:58 -0400
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <y0jiu6edoa9.fsf@vier.idi.ntnu.no>
Message-ID: <xj0d7wl9acd.fsf@cyclone.harvard.edu>

mlh at idt.ntnu.no (Magnus L. Hetland) writes:

> result = []
> for i in range(len(l)):
>     element = random.choice(list)
>     l.remove(element)
>     result.append(element)
> 
> Probably not that quick, but...

Indeed not that quick.  This shuffle is O(N**2) because each time
through the loop it has to (1) search for the element by value and (2)
delete the element which involved shifting the rest of the list over
by one.  Each of those actions is O(N) so the whole loop is O(N**2).

Michael

-- 
Michael Haggerty
mhagger at blizzard.harvard.edu



From nascheme at ucalgary.ca  Mon Aug 23 19:38:52 1999
From: nascheme at ucalgary.ca (Neil Schemenauer)
Date: 23 Aug 1999 23:38:52 GMT
Subject: gcc optimization breaks NumPy?
References: <7psar7$kt1$1@nnrp1.deja.com>
Message-ID: <7psm2c$ktk@ds2.acs.ucalgary.ca>

John Fisher <jfisher at are.berkeley.edu> wrote:
>I was getting consistent but inexplicable segmentation faults with the
>following code, using NumPy's C API to move data between Python and C
>(for sparse matrix multiplication; the data structure is from Meschach,
>in case that's relevant).  I found the problem to be caused by gcc's
>optimization options.

Which version of gcc?  With the lastest versions the strict
aliasing option is known to break some non-ANSI code.


    Neil



From yerny at primenet.com  Tue Aug 17 13:06:35 1999
From: yerny at primenet.com (Bryan Jernigan)
Date: Tue, 17 Aug 1999 10:06:35 -0700
Subject: CVS classes in Python
Message-ID: <37B9969B.BB40C25F@primenet.com>

I'm a newbie to Python and I would like to create a Python script to
maintain a distributed CVS CM system.  Before I re-invent the wheel, I
was wondering if anyone has some decent CVS classes and/or functions.
If so, please drop me a line.

Thanks for any info,
Bryan




From jblaine at shell2.shore.net  Thu Aug 19 11:29:37 1999
From: jblaine at shell2.shore.net (Jeff Blaine)
Date: Thu, 19 Aug 1999 15:29:37 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <slrn7ro8n1.h6v.jblaine@shell2.shore.net>

>GOTCHA: (low)
>    You can't use "in" on dicts.  Instead, you must use 
>	d = { "fred":"wilma", "barney":"betty" }
>	if d.has_key("fred"):	# legal
>	if "fred" in d:	        # ILLEGAL
>    I don't understand why this was done.

I'm very curious what would you think should be legal.

Would 'in' (when used on a dict) search the keys, values, or both?

Each one of those seem like a bad idea to me.  You have 2 separate
areas containing data.  Searching each of those individually 
(has_key()...) seems right to me.

Thanks for the document.  It was very interesting reading.



From martin at ics.mq.edu.au  Tue Aug 24 22:10:19 1999
From: martin at ics.mq.edu.au (Martin Ellison)
Date: Wed, 25 Aug 1999 12:10:19 +1000
Subject: word1.word2 notation
References: <oxgbtbxejea.fsf@panix7.panix.com>
Message-ID: <7pviui$dvv$1@sunb.ocs.mq.edu.au>

Yeoh Yiu wrote in message ...
>I've noticed in various languages, there is a construction of
expressions with a dot in the middle...
There are also some constructions with a "::" (C++), a "'" (Ada)
and a "!" (Visual Basic). These are similar that they provide a
kind of attribute, but with some sort of variation. As far as I
can remember:
C++ ns::a provides an identifier in a namespace (i.e. the "::" is
resolved at compile time).
Ada x'max [I think] provides a static attribute of "x" such as
the largest value it can take (i.e. the "'" is also resolved at
compile time).
Visual Basic rs!v is used for (among other things?) accessing a
value retrieved from a database. rs.EOF would provide an
attribute of the whole read (whether it is at end).
Hope this helps.

Martin






From nascheme at ucalgary.ca  Sat Aug 21 14:50:26 1999
From: nascheme at ucalgary.ca (Neil Schemenauer)
Date: 21 Aug 1999 18:50:26 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <7ple6r$kf0@ds2.acs.ucalgary.ca> <37bec222@cs.colorado.edu>
Message-ID: <7pmsdi$81u@ds2.acs.ucalgary.ca>

I think you misunderstand the intention of my post.  I am not
trying to defend Python on point.  Python is not perfect.  I am
trying to give some information of why things are the way they
are and am trying to help a new user of the language understand
it.

Is it really productive to keep repeating that length one tuples
in Python are weird?

Tom Christiansen <tchrist at mox.perl.com> wrote:
>In comp.lang.python, 
>    nascheme at ucalgary.ca (Neil Schemenauer) writes:
>:It is not too hard to remember what is mutable and what is not.
>
>I'm not sure that this maps well to the way people think, considering
>that everyone keeps telling me that it's "just something you have to
>learn and/or get used to", which means it's not intuitive at first.
>And that was my point of listing it as a gotcha.

I don't deny that this is a gotcha.

> Why can python copy sequences of chars (1-byte strings)
>trivially:
>
>    s = "this stuff"
>    t = s
>
>But as soon as you have a sequence of, oh, integers for example, you
>can't use the same set-up.  Very odd.

You are still confused.  Python does not copy anything
implicitly.  Read my other post about this.  Maybe the id()
function will help:

    >>> a = 'hello'
    >>> id(a)
    135476992
    >>> b = a
    >>> id(b)
    135476992
    >>>

[on perl copying objects]
>Yes, with 1st-class arrays (you say lists) and hashes (you say
>dictoinaries).  But not with instances, which are always by definition
>references, and suffer the same confusion.  That's why base types
>are easier to use for many "accidental programmers".

What happends when you have a large array and pass it to a
function?  This seems pretty inefficient.  If the array is not
copied then why are they so many rules about when something is
copied and when it is not.

>:>    Slices in Python must be contiguous ranges of a sequence.
>:>    In Perl, there's no such restriction.  
>:Use a dictionary if you want this.
>
>Huh?  How would I use a dictionary

Sorry, I didn't read you problem close enough.

[on del as a statement]
>I *really* don't undertsand this. Why does Python have so many
>"statements" compared with C or Perl?  What does this actually buy you?
>I can remove variable bindings just fine with the undef() or delete()
>functions in Perl, without needing to sacrifice an entire *statement*
>in the grammar for this!  There must be something important that I'm
>not understanding here, and I'd like to.

That undef() must be an interesting function to be able to modify
the binding of its caller.  You don't see a conceptual problem
with this?  Can user written functions do this too?


    Neil



From jdnier at execpc.com  Mon Aug  2 00:58:33 1999
From: jdnier at execpc.com (David Niergarth)
Date: Sun, 1 Aug 1999 23:58:33 -0500
Subject: Why is __init__.py required
References: <7o2eeq$gs2@dfw-ixnews6.ix.netcom.com>
Message-ID: <7o38g3$4iv@newsops.execpc.com>

G. David Kuhlman <dkuhlman at netcom.com> wrote in message
news:7o2eeq$gs2 at dfw-ixnews6.ix.netcom.com...
> I have looked and looked for this.  Whenever I put an empty
> __init__.py in a package, I ask myself why I have to do this.  I
> read Guido's article on packages and it told me that I have to put
> an __init__.py in a directory in order for the directory to be
> recognized as a package.  But it doesn't say why.

I just noticed earlier today that you can put code *in* an __init__.py file,
which makes that file seem a little more useful! For example, see the source
of mxTextTools ( http://starship.python.net/~lemburg/mxTextTools.html ) or
the XML package (and probably lots of other packages). It's a feature I've
never seen explained.

--David Niergarth






From stephan at pcrm.win.tue.nl  Mon Aug 23 03:03:22 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 23 Aug 1999 09:03:22 +0200
Subject: Unsorting(randomizing) a sequence
References: <000401beebac$7a89b520$0c2d2399@tim>
Message-ID: <m3d7wfm11h.fsf@pcrm.win.tue.nl>

"Tim Peters" <tim_one at email.msn.com> writes:

> [Tim, on the difficulty of generating random permutations, because
>  n! quickly grows larger than a given random-# generator's period]
> 
> [Aahz Maruch]
> > What happens if you re-seed?
> 
> By what means?  If you have a good source of random bits to do the reseeding
> with, save yourself the expense of calling random() and just use them
> directly <0.1 wink>. 

Actually, on Linux you could reseed by reading values from /dev/random.
/dev/random will block if not enough "random" events are happening
in the system to satisfy your application's hunger for random number.
So this might actually make a little sense; a *real* random source
(like /dev/random supposedly is) might be too slow to produce
enough numbers.

...although there is also /dev/urandom, which basically solves this
problem for you (i.e. it uses the same mechanism as /dev/random to create
random numbers, until the kernel is out of random numbers, and then
it continues with pseudorandom numbers).

Greetings,

Stephan



From fredrik at pythonware.com  Tue Aug 17 03:23:06 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Tue, 17 Aug 1999 07:23:06 GMT
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org>
Message-ID: <00c301bee884$42b1d3b0$f29b12c2@secret.pythonware.com>

Dan Connolly <connolly at w3.org> wrote:
> I have thought this was broken for years, but I just read
> it again tonight, and I'm not going to pass over it in silence this
> time
> 
> "Do not import this module directly. Instead, import the module os,
> which provides a portable version of this interface."
> -- http://www.python.org/doc/1.5.2p1/lib/module-posix.html
> 
> I think this is terrible advice.
> 
> The os module is *not* portable:

in fact, the "os" module docs says:

    This module provides a more portable way of using
    operating system (OS) dependent functionality than
    importing an OS dependent built-in module like posix
    or nt.

note the use of "more" in that sentence.

> If an implementation on WinNT provides posix semantics,
> then it may supply a module by that name that works as
> expected.

how about partial posix semantics? (like Windows offers,
which happens to be the most widespread target platform
we have today). hey, maybe the best solution is to offer
that under a different name?  (hmm, isn't that how it's
done today?)

and how to deal with optional features? like chown, exec*,
fork, ftruncate, getcwd, getegid, geteuid, getgid, getpgrp,
getppid, getuid, kill, link, mkfifo, nice, pipe, plock, popen,
putenv, readlink, setgid, setpgid, setpgrp, setsid, setuid,
strerror, symlink, system, tcgetpgrp, tcsetpgrp, times,
uname, wait, and waitpid, one or more of which are not
supported by all "mostly posixish" unix flavours.  hey, maybe
the best solution is to offer some way to test which of these
are available?  (wait, isn't that the case today?)

> Otherwise, I want an early ImportError, not
> later AttributeErrors, or worse: silent failures to
> behave as specified.

what about:

    import os
    
    # platform profile: list functions required by this
    # module, so we get early errors
    os.getpgrp, os.foo, os.bar

    ...

> It would have made sense for python to define a really portable os
> module, i.e. the least common denominator of the platforms it
> runs on. But that's not what we've got.
> 
> I hope the os module will be deprecated in future releases.

umm, so if it doesn't implement 100% of what YOU happen
to want from it, NOBODY should be allowed to use it.  that's
cool, but maybe somewhat impractical for the
rest of us ;-)

(fwiw, while I'm not all that surprised to here this from
people in your domain <0.5 wink>, history has showed
that it's a doomed approach:

 http://www.jwz.org/worse-is-better.html

)

</F>




From gmcm at hypernet.com  Fri Aug 20 01:27:49 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 20 Aug 1999 00:27:49 -0500
Subject: Win32 Installer - new beta
In-Reply-To: <37BB1373.6686F844@aw.sgi.com>
Message-ID: <1276999047-22443335@hypernet.com>

Gary Herron wrote:
> Gordon McMillan wrote:
> > 
> >  A new beta (02) of the Win32 Installer package is available on
> > starship:
...
> Nice work! 

Thank you.

> First the suggestion, then (in case you care) the description of
> what I'd use it for.
> 
> There appears to be two differences between Run.exe and Runw.exe:
>   (1)  Run.exe is a console app, and Runw.exe is not.
>   (2)  Run.exe cleans up all its extracted files, and Runw.exe does
>   not.
...
> Is there a good reason why Runw.exe does not cleanup after itself?

Sheer (stupid) oversight on my part.

I'm unburying myself from a client "emergency" (read lombotomized 
manangers assuaging existential insecurities by throwing tantrums). 

I see Thomas Heller pointed you to his version. He shipped them to me 
just about the time the "emergency" broke, so I haven't gone through 
them. I'm glad to see you've been testing them! They appear to take 
care of about half of what's been on my todo list.

- Gordon



From greg.ewing at compaq.com  Tue Aug 31 22:46:03 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 01 Sep 1999 14:46:03 +1200
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net> <99082819072100.03655@o112> <7q98g5$7bi$5@newshost.accu.uu.nl> <37C831BD.A5B17028@Lugoj.Com>
Message-ID: <37CC936B.DE4F9332@compaq.com>

James Logajan wrote:
> 
> I had nothing to do with the virtual death of Stallman. And
> he is really not dead;

Of course not, he's just resting...

Greg



From florent.heyworth at radbit.com  Sun Aug  1 21:36:08 1999
From: florent.heyworth at radbit.com (Florent Heyworth)
Date: Mon, 02 Aug 1999 03:36:08 +0200
Subject: RMI and Python
References: <Pine.GSO.4.10.9908011131490.25758-100000@xmission.xmission.com> <wkhfmj6vpd.fsf@ifi.uio.no>
Message-ID: <37A4F608.762989C5@radbit.com>


Lars Marius Garshol wrote:

> * Tim Egbert
> |
> | Java is able to do remote method invocation (RMI) which is a neat
> | concept.  It is the basis for Jini, another neat idea.  Does Python
> | have anything like RMI?
>
> Yes: CORBA. See <URL: http://www.dstc.edu.au/Fnorb/ >. I think there
> also are some bindings to various C ORBs floating around. Compared to
> RMI CORBA has the advantage that it is completely vendor-, language-
> and platform-independent.
>
> --Lars M.

As an alternative have a look at Dopy 0.2 at
http://www.users.cloud9.net/~proteus/dopy

to quote from the documentation:

" DOPY is a small distributed object system written entirely in Python.
It is not intended to be CORBA compliant. Instead, it aims to be
extremely easy to use and to support Python's dynamic nature - methods
are invoked dynamically, parameters are passed by copy. Any python object
that can be pickled can automatically be passed as a parameter or a
return value, and any Python object can be published as a distributed
object."

I've been playing around with it and I find it quite usable (see author's
notes)

Cheers

Florent Heyworth





From abigail at delanet.com  Sun Aug 15 17:40:06 1999
From: abigail at delanet.com (Abigail)
Date: 15 Aug 1999 16:40:06 -0500
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net> <slrn7re808.77t.kaz@ashi.FootPrints.net>
Message-ID: <slrn7recvk.81j.abigail@alexandra.delanet.com>

Kaz Kylheku (kaz at ashi.FootPrints.net) wrote on MMCLXXV September MCMXCIII
in <URL:news:slrn7re808.77t.kaz at ashi.FootPrints.net>:
## On 15 Aug 1999 03:14:48 GMT, Ben Caradoc-Davies <bmcd at es.co.nz> wrote:
## >[strange programming]
## >The most confidence uninspiring piece of code I ever had the displeasure to
## >maintain contained something like this (allegedly C++, but mainly it's 
## >intersection with C):
## >
## >	i = 0;
## >	while( i < max_index ) {
## >		/* some copying or update (not modifying i) goes here */
## >		i = i + 1;
## >	}
## >
## >This was written by a *very* expensive consultant working for a large
## >multinational accountancy firm.
## 
## So what's wrong with that? Granted, the for loop construct would have
## been better for the task instead of while, and the expensive consultant shoul
## have seen that. But what about the i = i + 1 increment? It smacks of BASIC, b
## there is nothing wrong with it.  It's not any less efficient than i++ or ++i.


I would still argue we don't know enough about the program to say it's
wrong or bad style. The last line is easily modified to "i = i + 2", or
"i = i * 1.5", or "i = i + a", where a is depending on the rest of the
statements in the block. I would prefer using "i = i + 1" or "i += 1"
if I mean "I add a number to i, which happens to be 1 in this case" over
"i ++", which in my book means, "I am counting" or "it's the next one".
I don't think using 'for (i = 0; i < max_index; i = i + 1) {}' is going
to buy me anything over the while. Except that's less flexible. Thinking
about the next programmer? Gimme a break. A programmer that gets confused
when seeing a while that could be written as a for shouldn't touch the
code anyway.


It's the same with !i and i == 0. With !i, i is acting as a boolean, while
in "i == 0", I am comparing i to a specific value. That the value just
happens to be 0 doesn't warrent a change in syntax.


Abigail
-- 
$" = "/"; split $, => eval join "+" => 1 .. 7;
*{"@_"} = sub  {foreach (sort keys %_) {print "$_ $_{$_} "}};
%{"@_"} = %_ = (Just => another => Perl => Hacker); &{%{%_}};


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From roy at popmail.med.nyu.edu  Sat Aug 21 14:25:26 1999
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Sat, 21 Aug 1999 14:25:26 -0400
Subject: Man pages would be cool.
References: <37BC6986.A66B044E@starmedia.net> <37BE24E0.79EA064E@montana.com> <37BE9EAB.CB7B1380@prescod.net> <37BEDBF3.BCA64E58@montana.com> <7pmqgq$354ai@fido.engr.sgi.com>
Message-ID: <roy-2108991425270001@mc-as01-p02.med.nyu.edu>

pj at sgi.com (Paul Jackson) wrote:
> I recall for 6 months when I was bringing up a 68010 Unix box, I'd just
> reboot the lab machine anytime I found it left in vi by another engineer,
> because I didn't know how to quit vi.

The only vi I know is "<esc>:q!<return>".  So far, it's been all the vi
I've ever had to know :-)  If I ever have to work on a machine that
doesn't have emacs, I just use ed.



From kikutani at sprintmail.com  Wed Aug 25 13:55:08 1999
From: kikutani at sprintmail.com (Kikutani Makoto)
Date: 25 Aug 1999 17:55:08 GMT
Subject: PyKDE for Win ?
Message-ID: <7q1als$id9$1@ash.prod.itd.earthlink.net>

Looks like a ridiculous qestion, but is there any plan to
port PyKDE to Win world ?

-- 
Kikutani, Makoto  kikutani at sprintmail.com




From shecter at darmstadt.gmd.de  Thu Aug 19 11:25:13 1999
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Thu, 19 Aug 1999 17:25:13 +0200
Subject: Kudos to documentors: (popen2)
References: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de> <37bc15c2@cs.colorado.edu>
Message-ID: <37BC21D9.E19DBEA2@darmstadt.gmd.de>

Hi,

I really didn't intend to attack you in a back-door kind of way.  I
actually first thought about posting this to a perl group to show the
problems I see.  I guess my decision was colored by negative previous
experiences I've had with strong perl-fans who love the perl system
and perldoc just as it is.  

That was probably a mistake of mine to not post in a perl group.  

Anyhow, to my comments - part of the problem here is also the perl
language, which exagerates any problems in less than perfect
documentation.  So, I know that what I'm going to say may not have
100% to do with the docs themselves:


Tom Christiansen wrote:
>
> >http://www.perl.com/CPAN/doc/manual/html/lib/IPC/Open2.html
> 
> The standard documentation for Perl is found in manpages, not
> on the web. 

It ought to be.  Or at least, well-linked "info" pages.  (This is
surely easily agreed upon, right?)  But this is a moot issue, because:

> 
>     man IPC::Open2
> 

...has apparently the same text that bothered me.  


> 
> :(Just notice how much is implicit, and not documented, and with no
> :links to other needed info.)
> 
> I have no idea what you're talking about, or what links you're
> looking for, but as the original author of that module, I am
> always happy to accept documentation suggestions.

Well, you really have to put yourself in the shoes of someone who uses
perl sometimes, even frequently, but can't always remember all of
perl's (very numerous) special cases, symbols, and objects:

1)  It will not create these file handles for you.  You have
       to do this yourself.  So don't pass it empty variables
       expecting them to get filled in for you. 

My question: how do I create these file handles?  What document would
I look in to find out?  (Aside: this to me is a case of the complex
perl language aggrevating the work of the documentor: The user must
understand "file handles", a sort of special perl-entity, as well as
cope with a lib designed with in-out parameters.  (???)  Compare the
Python version for ease of use.)

2) If $rdr is a string (that is, a bareword filehandle rather
       than a glob or a reference) and it begins with ">&"...

Woof!  Complicated!  I would have liked to use the feature talked
about (if I understand it correctly), but now there's another special
entity discussed, which I also don't know how to create.  A "bareword
filehandle"?  I wanted to hook up stdin and stdout... how do I get a
bareword filehandle out of one of those?  (It's ok - I don't need the
answers to these anymore - I'm using Python, but these are the
problems I had.)


> Or is this just another case of pissing on something that you didn't
> understand out of the way of the people responsible?  If so, you lose,
> because *I'm* responsible.  What's your biggest problem?

No; that document was really the item that sent me to Python.  Sure,
it was the straw that broke the camel's back (Hmm!) - I had been
fighting with Perl for a good hour or so, getting "irrelevant" error
messages, and then seeing docs that posed more questions for me than
they answered caused me to give Python a try.

BTW, I did appreciate your notes about deadlocks.  That -was- easy to
understand for the typical experienced unix programmer.  


> 
> And remember, Perl's documentation is NOT WEB-BASED.

OK - got it.  But how do you feel about that?  I think it ought to be
- it should take a look at what Javadoc and Python's doc system do
right, and do it better.  Ie: starting with semantic markup, not just
formatting markup.


> 
>            use FileHandle;
>            use IPC::Open2;
>            $pid = open2(*Reader, *Writer, "cat -u -n" );
>            Writer->autoflush(); # default here, actually
>            print Writer "stuff\n";
>            $got = <Reader>;

...This is even more confusing.  How are the filehandles being created
beforehand?  Is that part skipped?  Or do they not really need to be? 
Also, here I see that they're written like *Reader and not \*Reader as
the Open2 synopsis has it.  Not sure what that means.  (Again -  I
think that Perl's complexity sabotages the documentor... if everything
were simply objects, and also objects used in an OO way, then there
would be much less need to document.)


 
> Next time, please try the manpages, and send the authors mail or file a
> bug report.  But please don't just piss on people where you think they
> can't see you.  Sheesh.
> 

I certainly will.  Sorry,

- Robb



From gerrit.holl at pobox.com  Mon Aug 30 08:59:36 1999
From: gerrit.holl at pobox.com (Gerrit Holl)
Date: Mon, 30 Aug 1999 12:59:36 GMT
Subject: Upper limit?!
In-Reply-To: <al8aer9czzq.fsf@sirppi.helsinki.fi>; from Markus Stenberg on Mon, Aug 30, 1999 at 01:41:13PM +0300
References: <37CA1E0F.FA2A05A3@megsinet.net> <014f01bef2bd$3fd09e90$f29b12c2@secret.pythonware.com> <37CA3EBD.4077AF18@megsinet.net> <al8aer9czzq.fsf@sirppi.helsinki.fi>
Message-ID: <19990830145936.B6858@optiplex.palga.uucp>

On Mon, Aug 30, 1999 at 01:41:13PM +0300, Markus Stenberg wrote:
> David Oppenheimer <davidopp at megsinet.net> writes:
> > Wow, that's quite a difference! on an Alpha you can have a 19 digit number
> > as opposed to a 10 digit number on my Pentium.  So the limit isn't actually
> > a factor of the programming language but a virtue (or limitation) of my
> > processor.  I wonder if Intel knows about this?!   LOL  Makes me want to
> 
> Yes, for machine-spesific ints.. However, some languages can kludge around
> it, like Python's bigints:
> 
> fingon at dreamland ?uname -m
> i686
> fingon at dreamland ?python
> Python 1.5.2 (#0, Jul 16 1999, 19:02:32)  [GCC egcs-2.91.66 Debian
> GNU/Linux (egcs-1.1.2 release)] on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> len(str(2L**666))
> 202
> >>>
> 
> 202 digits, baby ;-)
> 

201, you shouldn't count the 'L' on the end ;-)

regards,
Gerrit.



From skip at mojam.com  Tue Aug 10 10:39:27 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 10 Aug 1999 09:39:27 -0500 (CDT)
Subject: Hidden CGI fields in HTML
In-Reply-To: <37A94EDA.81644EF3@NOSPAM.strs.co.uk>
References: <37A94EDA.81644EF3@NOSPAM.strs.co.uk>
Message-ID: <14256.14722.703130.225407@dolphin.mojam.com>

    Ian> I am using CGI with Python and am trying to find a way that I can
    Ian> generate the HTML for a form such that one of the INPUT controls
    Ian> will have a fixed value (set when the page is generated by a
    Ian> previous CGI script), that CANNOT be edited by the user (or
    Ian> prefereably not even seen) - but will be passed to the next CGI
    Ian> script.

<input type=hidden name=spam value=eggs>

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From robin at jessikat.demon.co.uk  Sat Aug  7 08:46:26 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Sat, 7 Aug 1999 13:46:26 +0100
Subject: Linux why not fully dynamic?
References: <1278177653-25589496@hypernet.com>
 <37AAF3D5.141B55EB@bioreason.com> <37AB003C.28D9@mailserver.hursley.ibm.com>
Message-ID: <Yvk34MAiqCr3Ewcx@jessikat.demon.co.uk>

In article <37AB003C.28D9 at mailserver.hursley.ibm.com>, Paul Duffin
<pduffin at mailserver.hursley.ibm.com> writes
...
>> 
>> (This is a Unix question; on Mac and Windows, it is a shared library.)
>> 
>> It's just a nightmare to get this to work on all different
>> platforms. Shared library portability is a pain. And yes, I know about
>> GNU libtool -- but it requires me to use its conventions for filenames
>> etc, and it would require a complete and utter rewrite of all the
>> makefile and config tools I'm currently using.
...
>> In practice, few applications embed Python -- it's much more common to
>> have Python extensions, which already are shared libraries. Also,
>> serious embedders often want total control over which Python version
>> and configuration they use so they wouldn't want to use a standard
>> shared library anyway. So while the motivation of saving space when
>> lots of apps embed Python is nice in theory, I doubt that it will save
>> much in practice.  (Hence the low priority I give to making a shared
>> library.)
>> -----------------------------------------------------------------------
The windows version uses DLLs and there is a mechanism already for
extensions to know what version of python they're built for. Surely most
unices are just as good as if not better than win32 when it comes to
preventing DLL hell.

My view on python is that much of the success in win32 comes from the
great work done by Mark Hammond & buddies which allows python to be used
easily as a com object generator/interfacer etc etc. Some of the
problems come from the fact that python15.dll doesn't work exactly the
same as expected ie it doesn't know where python.exe is etc. PythonCOM
makes use of DLL's for inproc servers.

Can CORBA and the like make use of python in the same kind of way and
would this be embedding?
>
>AFAIK this is a quote from GvR which was given a long time ago, I would
>be interested to know if he has changed his mind on this.
>
>Whenever anyone asks this question they "eventually" get pointed to this
>quote. I think that there have been enough people asking about this to
>invalidate the primary reason for this being given such a low profile.
>
>Considering the fact that Python already has support for dynamic
>extensions and that it uses backlinking which does not care whether
>the symbols are in a library or an executable it is not as difficult
>as Guido thinks.
>
>Unfortunately because of this statement people are reluctant to do
>the necessary work to support this because they do not feel that it
>will be added into the code. Instead they workaround it or do enough
>for themselves and so the problem never gets fixed.
>
Paul was a major force behind the stubs mechanism in Tcl (I believe
because of failings in AIX). The stubs mechanism allows versions to be
more independent and overcomes deficiencies in inter extension linking.
There is certainly less of this in python as extensions tend to
implement specific python objects and normally don't refer across
directly to the internals of other extensions.

However, I see no reason why python shouldn't be split into smaller bits
if only to reduce the footprint. The parser module is unwanted for many
applications and certainly under Linux the executable size came down by
40% when the *shared* Setup option was used.
-- 
Robin Becker



From tim_one at email.msn.com  Tue Aug 17 03:47:10 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 03:47:10 -0400
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <7pb1eb$elt$1@nnrp1.deja.com>
Message-ID: <000401bee884$b6196fc0$112d2399@tim>

[Ole-Kristian Villaboe]
> Is there a command to do this?

No.

> ...
> I've been writing an algorithm myself, but it's incredibly slow and I
> would like something quicker (of course :) )

Post your algorithm; can't say how to speed it up without knowing why it's
incredibly slow; for that matter, can't guess what "incredibly" means.
Perhaps quadratic-time in the number of elements?

no-info-in-no-info-out-ly y'rs  - tim





From loewis at informatik.hu-berlin.de  Mon Aug 16 04:55:01 1999
From: loewis at informatik.hu-berlin.de (Martin von Loewis)
Date: 16 Aug 1999 10:55:01 +0200
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au> <7p7aaj$sdl$1@cronkite.cc.uga.edu>
Message-ID: <p6qhfm0nlzu.fsf@informatik.hu-berlin.de>

graham at sloth.math.uga.edu (Graham Matthews) writes:

> : 	s = ""
> : 	for  i in range(1,10): s = s + str(i)
> : 
> : Obviously i is an integer here, and s a string.
> 
> Is it? Surely that depends on what "range" and "+" are bound to at
> runtime.

He said that in Viper, range is a keyword. How could you bind "+" to
something else in the example?

Regards,
Martin



From tim_one at email.msn.com  Sat Aug 28 04:11:57 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 28 Aug 1999 04:11:57 -0400
Subject: Why use Perl when we've got Python?!
In-Reply-To: <935793922.672565@fezzik.endicor.com>
Message-ID: <000801bef12c$fef6e140$ad2d153f@tim>

[Kent Polk]
> I believe you are thinking of Al'Khwarizmi, who wrote the book
> "Hisab al-jabr w'al-muqabala". 'Algebra' came from 'al-jabr' and
> 'Algorithm' came from the name Al'Khwarizmi.

Nonsense.  Everyone knows 'Algorithm' came from Al Gore, Vice President of the
United States and Inventor of the Internet.

> Later Greeks certainly appear to have used the concept of
> variables, but did have not the algebraic notation.

Well, at least you got *that* part right <wink>.

and-john-backus-is-the-god-of-wine-or-something-like-that-ly y'rs  - tim





From moritz at news.uni-mannheim.de  Thu Aug 26 04:33:19 1999
From: moritz at news.uni-mannheim.de (Moritz Moeller-Herrmann)
Date: Thu, 26 Aug 1999 10:33:19 +0200
Subject: PyKDE for Win ?
References: <7q1als$id9$1@ash.prod.itd.earthlink.net> <37c45fb2.185911967@midpoint>
Message-ID: <slrn7s9uuf.q2.moritz@Gertrud.chaos.wg>

On Wed, 25 Aug 1999 21:28:13 GMT, Timothy Dixon <tdixon.no at spam.fwi.com> wrote:
>On 25 Aug 1999 17:55:08 GMT, kikutani at sprintmail.com (Kikutani Makoto)
>wrote:

>>Looks like a ridiculous qestion, but is there any plan to
>>port PyKDE to Win world ?

>Um.... Unless I'm wrong (which has happened a nonzero number of
>times), PyKDE is bindings to the KDE Window Manager in XFree86.  Since
>KDE doesn't exist in the Windows world, what good would PyKDE do?

KDE is not a window manager. KDE is a desktop environment besed on the QT
widget set. It works with any X and on most unices.

pykde is (AFAIK) wrapping around both the qt widgets and the added KDE
widgets.

As the QT widgets are cross-platform, it might not be too hard to port them to
windows (might cause licensing problems though...)
KDE on the other hand assumes a real unix operating system.

-- 
Moritz Moeller-Herrmann mmh at gmx.net ICQ# 3585990        # Not only
Get my public pgp / gpg key from                        # Open Source(TM)
http://webrum.uni-mannheim.de/jura/moritz/pubkeymoritz  # but also
KDE forever! Use Linux to impress your friends!         # Open Minded!




From jstevens at bamboo.verinet.com  Sat Aug 14 11:30:10 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: Sat, 14 Aug 1999 15:30:10 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cf5f@cs.colorado.edu> <7p2lqn$2pf@dfw-ixnews7.ix.netcom.com>
Message-ID: <slrn7rb2s2.cf9.jstevens@bamboo.verinet.com>

On 14 Aug 1999 02:55:19 GMT, Eric Bohlman <ebohlman at netcom.com> wrote:
>John W. Stevens (jstevens at basho.fc.hp.com) wrote:
>[quoting someone whose name was lost]
>: > It is wrong from the Larry's principle that things that behave
>: > differently should have different appearances.
>: 
>: Then why doesn't Perl have two different operations for
>: assign/reassign, vs. add to an array?  A tiny little violation of
>: Larry's principles, that.
>
>Nope.  "Behave differently" hear means "accomplish two different tasks."  
>Comparing two strings to see if they're equal lexically is not the same 
>task as comparing two strings to see if they're equal numerically.  

Right.  Do you see what I am saying?  Automatic coercion (from string to int)
created a problem.  The problem was dealt with by making a change that
impacts the language.

As well as providing a huge opportunity for defects, the complexity
of the language was increased.

>Putting a particular value in a particular position in an array that 
>already holds a value *is* the same task as putting a particular value in 
>a particular position of an array that does not yet hold a value; the 
>only difference is in the internal details of how to accomplish it.

Yes, but the context of this was TC's comparison with Python's lists.

For an ARRAY, your argument makes sense.  For a list, it does not,
yet TC was taking Python lists to task for not acting like arrays.

On the other hand, Perl's behavior encourages defects:

$array[$index] = $newWord;

The program ran for a very long time, then the system started thrashing,
then it core dumped.

The value of $index?  Very, very large.  Perl happily tried to allocate
Gigabytes worth of core.

>Of course, the OOically-correct way to handle the string comparisons is 
>to define a base class, ComparableString, and two subclasses, 
>LexicallyComparableSting and NumericallyComparableString, along with 
>polymorphic operators that understand which class they're comparing to 
>and apply the correct comparison method.  But there are cases where you 
>may want to do *both* comparisons on a string,

I would disagree with the last part . . . it makes sense to encode
a number as an ASCII string, then to decode it back to an integer,
but to compare two strings as if they were numbers, again, allows
for far to many defects.

For example (another real bug, here):

$result = $Total / $count;

Generated a division error.  $count contained the string: "Payroll". . .

>and to be OC you then have 
>to cast the string to the appropriate type;

Cast?

>the Perlish way of picking 
>the operator director is IMHO easier and more maintainable; in my 
>experience, heavy casting makes code unreadable.

Casting is a poor idea, I agree.  Explicit conversion, however, is
more maintainable and much easier to use in a way that reduces
defect rates.

John S.



From john.deford at staarinc.com  Tue Aug 24 17:16:44 1999
From: john.deford at staarinc.com (John DeFord)
Date: Tue, 24 Aug 1999 16:16:44 -0500
Subject: Embedding Python into an MFC app - thread problems.
Message-ID: <7pv1jd$gmd@newsops.execpc.com>

I've embedded Python + win32 extensions into my MFC app with the following
code scrap in my app's InitInstance member (taken from Pythonwin embedding
documentation):

 Py_Initialize();

 glue.bShouldAbandonThreadState = false;  // Wasn't in example!

 if (!glue.DynamicApplicationInit("import initscore", NULL))

 // Assuming you have a ReportError method - do whatever makes sense!
  AfxMessageBox("Could not attach to the Python win32ui extensions.",
MB_OK);
  return FALSE;
 }

ptPythonThread = PyThreadState_Swap(NULL); // Wasn't in example!

I'm using MSVC++ 6.0, and my application is responsible for displaying all
windows at the moment.  The two lines commented by "// Wasn't in example",
plus the following lines bracketing subsequent Python calls seemed to be
required to get it to work:

  PyThreadState* ptThread = ((CCtest2App*)AfxGetApp())->ptPythonThread; //
Stored by call in InitInstance.
  PyThreadState* _save = PyThreadState_Swap(ptThread);

....// various Python C api function calls...

  PyThreadState_Swap(NULL);

Without all this thread swapping Python throws a fatal error saying current
thread is NULL.  Am I missing something, or is this additional code really
necessary?  The sources I'm working with are all the August 1999 versions.

John






From akuchlin at mems-exchange.org  Tue Aug 31 17:50:06 1999
From: akuchlin at mems-exchange.org (Andrew M. Kuchling)
Date: 31 Aug 1999 17:50:06 -0400
Subject: lex / yacc for Python
References: <7qfvct$o8$1@news.is.co.za>
Message-ID: <3d4shfiprl.fsf@amarok.cnri.reston.va.us>

"Jacques Oosthuizen" <jacques at integra.co.za> writes:
> Any one know if there is such a beast

Quite a few, actually; many Python users seem to be language geeks.  
A list of parsing systems is at:

http://starship.python.net/crew/amk/python/string.html#parsing

I've never used any of them, and don't know which ones are still
maintained, which are toys, and which are the most powerful.  Good luck!

-- 
A.M. Kuchling			http://starship.python.net/crew/amk/
This isn't life in the fast lane, it's life in the oncoming traffic.
    -- Terry Pratchett, in alt.fan.pratchett





From philh at vision25.demon.co.uk  Tue Aug 10 18:43:23 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 10 Aug 99 22:43:23 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <19990805094020.A1838753@vislab.epa.gov> <37aacdf9.324492916@news.oh.verio.com> <934117874snz@vision25.demon.co.uk> <7om59k$cqc$1@newshost.accu.uu.nl> <934237862snz@vision25.demon.co.uk> <7oop8f$hre$1@newshost.accu.uu.nl>
Message-ID: <934325003snz@vision25.demon.co.uk>

In article <7oop8f$hre$1 at newshost.accu.uu.nl>
           m.faassen at vet.uu.nl "Martijn Faassen" writes:
> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> I see, that makes sense, but you do lose all the work that's being put into
> XML, and all the hype, and standard Python libraries to deal with XML, etc.
> 
> Perhaps you can make it so that your abbreviated format can be readily
> translated to XML?

With the code I've currently written, it would be very easy to write
a translation to XML. Going the opposite way (XML -> my internal 
structure) would presumably also be easy given that there are XML 
parsers available for Python.
 
> > It will also use the C preprocessor, so for example one could
> > write a macro for a label and associated textfield:
> 
> > #define labelTF(txt,id,size) \
> >    label @id##Label txt textField @id cols=size
> 
> You'd need a separate XML preprocessor to do something like this (or 
> again just use XML for the backend). I'll stop harping on the XML for now,
> I was just wondering; it would make it easier for people to pick up this
> GUI definition language, and it seems applicable.

There's a Java-based GUI builder which uses an XML-based format
(JUML I think it is called). I have seen what XML files for this look
like, and IMO they are unwieldy, ugly-looking and too big.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From sca at isogmbh.de  Fri Aug  6 07:54:57 1999
From: sca at isogmbh.de (Chris...)
Date: 6 Aug 1999 11:54:57 GMT
Subject: Problems with Grail 0.6 and Python 1.5.2 (NT)
Message-ID: <933940498.641759@mhs0096x.mchh.siemens.de>

Hello

  I've just downloaded the latest version of Grail, but got the following
error:
-+--+--+--+--+--+--+ schnippel +--+--+--+--+--+--+--+--+--+--+--+--+-- 
C:\TEMP\Grail\grail-0.6>grail.py
Traceback (innermost last):
  File "C:\TEMP\Grail\grail-0.6\grail.py", line 499, in ?
    main()
  File "C:\TEMP\Grail\grail-0.6\grail.py", line 155, in main
    browser.home_command()
  File "C:\TEMP\Grail\grail-0.6\Browser.py", line 400, in home_command
    self.context.load(home)
  File "C:\TEMP\Grail\grail-0.6\Context.py", line 487, in load
    scrollpos=scrollpos)
  File "C:\TEMP\Grail\grail-0.6\Context.py", line 378, in read_page
    Reader(self, url, method, params, show_source, reload, data, scrollpos)
  File "C:\TEMP\Grail\grail-0.6\Reader.py", line 447, in __init__
    if url: self.restart(url)
  File "C:\TEMP\Grail\grail-0.6\Reader.py", line 491, in restart
    BaseReader.__init__(self, self.last_context, api)
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 62, in __init__
    self.start()
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 85, in start
    self.checkapi_regularly()
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 157, in checkapi_regularly
    self.callback()
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 184, in checkmeta
    self.getapimeta()
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 206, in getapimeta
    self.handle_meta(errcode, errmsg, headers)
  File "C:\TEMP\Grail\grail-0.6\Reader.py", line 544, in handle_meta
    if not self.handle_meta_prelim(errcode, errmsg, headers):
  File "C:\TEMP\Grail\grail-0.6\Reader.py", line 534, in handle_meta_prelim
    self.restart(url)
  File "C:\TEMP\Grail\grail-0.6\Reader.py", line 491, in restart
    BaseReader.__init__(self, self.last_context, api)
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 62, in __init__
    self.start()
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 85, in start
    self.checkapi_regularly()
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 157, in checkapi_regularly
    self.callback()
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 184, in checkmeta
    self.getapimeta()
  File "C:\TEMP\Grail\grail-0.6\BaseReader.py", line 206, in getapimeta
    self.handle_meta(errcode, errmsg, headers)
  File "C:\TEMP\Grail\grail-0.6\Reader.py", line 557, in handle_meta
    bkmks.record_visit(self.url, last_modified)
  File "C:\TEMP\Grail\grail-0.6\ancillary\BookmarksGUI.py", line 1000, in record
_visit
    bookmarks = self.get_bookmarks_by_uri(uri)
  File "C:\TEMP\Grail\grail-0.6\ancillary\BookmarksGUI.py", line 996, in get_boo
kmarks_by_uri
    self.initialize()
  File "C:\TEMP\Grail\grail-0.6\ancillary\BookmarksGUI.py", line 900, in initial
ize
    root, reader = self._iomgr.load(usedefault=1)
  File "C:\TEMP\Grail\grail-0.6\ancillary\BookmarksGUI.py", line 269, in load
    fp, reader = self.__open_url_for_reading(filename)
  File "C:\TEMP\Grail\grail-0.6\ancillary\BookmarksGUI.py", line 213, in __open_
url_for_reading
    fp = urllib.urlopen(url)
  File "C:\TEMP\Grail\grail-0.6\grail.py", line 344, in open_url_simple
    api = self.open_url(url, 'GET', {})
  File "C:\TEMP\Grail\grail-0.6\grail.py", line 339, in open_url
    api = self.url_cache.open(url, method, params, reload, data=data)
  File "C:\TEMP\Grail\grail-0.6\CacheMgr.py", line 159, in open
    return self.open_get(key, url, mode, params, reload, data)
  File "C:\TEMP\Grail\grail-0.6\CacheMgr.py", line 199, in open_get
    item = SharedItem(url, mode, params, self, key, data)
  File "C:\TEMP\Grail\grail-0.6\Cache.py", line 74, in __init__
    self.reset()
  File "C:\TEMP\Grail\grail-0.6\Cache.py", line 91, in reset
    data=self.postdata)
  File "C:\TEMP\Grail\grail-0.6\protocols\ProtocolAPI.py", line 119, in protocol
_access
    access = app.find_extension('protocols', sanitized).access
AttributeError: 'None' object has no attribute 'access'
-+--+--+--+--+--+--+ schnippel +--+--+--+--+--+--+--+--+--+--+--+--+-- 
My system is Windows NT 4 (SP5), Python 1.5.2 and TCL8 (that from the Python
distribution, btw. the installation program doesn't add a path to tcl
automatically, that was my first problem...).

  What's going wrong here (Grail 0.5 did work alright).

  Thanks a lot in advance!

bye
  Chris...



From jstevens at basho.fc.hp.com  Fri Aug 13 22:53:46 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 19:53:46 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132026.OAA03870@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:26 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cc2a@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :It comes down to the issue we discussed once before: Python is the
> :better language for [YOUR NEWSREADER SCREWED THIS UP]
> :OO, and I write OO most of the time.  I know you disagree.  But my
> :students, [YOUR NEWSREADER SCREWED THIS UP]
> :in general, also believe that Python is a better OO scripting language
> :than [YOUR NEWSREADER SCREWED THIS UP]
> :Perl.
> 
> What's a "scripting" language, for goodness sake?

A scripting language?  Read Ousterhout's paper on the subject.

Admittedly, the classifications are not sharply bounded.

John S.




From skaller at maxtal.com.au  Fri Aug 20 17:53:59 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Fri, 20 Aug 1999 21:53:59 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <7orbpq$i4s$1@newshost.accu.uu.nl> <37b6dea0.342930367@news.triode.net.au> <7p7aaj$sdl$1@cronkite.cc.uga.edu> <p6qhfm0nlzu.fsf@informatik.hu-berlin.de> <7p963d$t9$1@cronkite.cc.uga.edu>
Message-ID: <37bdcd5b.797261811@news.triode.net.au>

On 16 Aug 1999 14:09:49 GMT, graham at sloth.math.uga.edu (Graham Matthews) wrote:

>graham at sloth.math.uga.edu (Graham Matthews) writes:
>: > : 	s = ""
>: > : 	for  i in range(1,10): s = s + str(i)
>: > : 
>: > : Obviously i is an integer here, and s a string.
>: > 
>: > Is it? Surely that depends on what "range" and "+" are bound to at
>: > runtime.
>Martin von Loewis (loewis at informatik.hu-berlin.de) wrote:
>: He said that in Viper, range is a keyword. How could you bind "+" to
>: something else in the example?
>
>You might have bound it in some code previous to this. 

	It is possible to detect if this might be the case,
provided we exclude rebinding 'range' in an 'exec'.
And, since most people don't rebind range,
assuming it isn't is probably reasonable.

>If you are
>planning to compile this code then you have to compile it so that it
>will run in all contexts.

	No, I don't. I plan to compile _whole_ programs.
The result of compiling a function 'f' in some module 'm'
could be different in different programs. In fact,
it could be different for different _calls_: there could be
two or more versions of one function, depending on
usage.

>What I would like to know here is what is the type of range(1,10)?

	At present, it has type 'IntRange' which is analogous
to Python 1.5.2's type 'XRange': a dictinct type which is a sequence.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From Ted.Horst at wdr.com  Thu Aug 26 14:22:27 1999
From: Ted.Horst at wdr.com (Ted Horst)
Date: Thu, 26 Aug 99 13:22:27 -0500
Subject: fully qualified hostname ?
Message-ID: <199908261822.AA01099@ch1d2833nwk>

What is the best way to get the fully qualified hostname in python (ie  
hostname.domainname) ?

A portable solution would be ideal, but I would settle for Unix only.

Thanks,

Ted
---



From kistler at fnmail.com  Fri Aug 27 16:26:52 1999
From: kistler at fnmail.com (Per Kistler)
Date: Fri, 27 Aug 1999 22:26:52 +0200
Subject: Python in Blender
Message-ID: <37C6F48C.4C465AFB@fnmail.com>

Hi

Someone checked out Python in Blender already?
Is seems to be a good idea, to have Python as an
builtin language in a modeling/rendering tool.

-Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From eye at niksula.hut.fi  Fri Aug 20 14:01:37 1999
From: eye at niksula.hut.fi (Petri Mikael Kuittinen)
Date: 20 Aug 1999 21:01:37 +0300
Subject: How to sell Python to your boss?
References: <ka8yaf7d84t.fsf@delightful.cs.hut.fi>
Message-ID: <ka8vhaaz5z2.fsf@delightful.cs.hut.fi>

I wrote:

 > [snip]
 > My boss
 > otherwise agreed with my idea ("it would be really useful..."), but he
 > doesn't want me to do the prototype implementation using Python. He
 > said I must use a program language that all the other programmers in
 > our R&D department understand. Well, that leaves only one choice: C
 > [snip]

Well, it looks like it is Christmas. :-)

After posting this article I talked with one of my colleagues about
the future of software development. Today he and one other colleague
went to talk with my boss about this issue and the result is that I
have been allowed use Python in this project. My boss is interested to
see what are the benefits of using Python compared to C.

Thanks everyone who suggested good ideas to sell Python!


Petri

-- 
<(O)> Petri Kuittinen, also known as Eye, Dj Eye or Peku               <(O)>
<(O)> ADDRESS: Timpurinkuja 1 C 39, FIN-02600 Espoo, Finland           <(O)>
<(O)> EMAIL: eye at iki.fi WWW: http://www.iki.fi/~eye/ PHONE: 09-5472380 <(O)>
~Vegetarians beware! You are what you eat.



From da at ski.org  Thu Aug 12 14:35:12 1999
From: da at ski.org (David Ascher)
Date: Thu, 12 Aug 1999 11:35:12 -0700 (Pacific Daylight Time)
Subject: Copying Huge Files ? (Newbie!)
In-Reply-To: <99081220382305.14681@o112>
Message-ID: <Pine.WNT.4.04.9908121133330.258-100000@rigoletto.ski.org>

On Thu, 12 Aug 1999, Fredrik Henbjork wrote:

> On Thu, 12 Aug 1999, you wrote:
> > 
> > I want to copy some files across the disks of my server using python
> > parse a script and then do what I want.
> > 
> > But while reading the manual I cannot find something like os.copyfile
> > or os.movefile.
> > 
> > Are there some functions except open/read/write that can do the copy
> > for me ???

> > Ciao
> > 	Matthias Barmeier
> 
> Take a look at the things in shutil (especially shutil.copy and
> shutil.copyfile) and os (especially os.rename and os.renames).

I'd like to point out that especially if the files are indeed huge, then
the operating system file copy operations may be *much* more efficient
than shutil.copy*, e.g. by using mmap when appropriate.

os.system() can be your friend...

--david




From doug at tippett.com  Thu Aug 26 20:28:54 1999
From: doug at tippett.com (Doug Epps)
Date: Thu, 26 Aug 1999 17:28:54 -0700
Subject: extension and core dumpage
Message-ID: <37C5DBC6.669869E7@tippett.com>

hi,

i'm trying to experiment with extending python with c/c++
and getting the following message on 'import '

Fatal Python error: PyThreadState_Get: no current thread
Abort (core dumped)


SGI IRIX 6.3

the extension code is really simple

any ideas or pointers would be great.

please email too, i dont trust our news server

doug

## here's the code
#include <Python.h>


static PyObject *ErrorObject;
extern "C" {
  void initdougTest ();
};

static PyObject *top (PyObject *, PyObject *) {
  Py_INCREF(Py_None);
  return (Py_None);
}

static struct PyMethodDef dougMethods[] ={
  { "top", top, 1, "the stuff"},
  {NULL, NULL, 0}
};
void initdougTest ()
{
  PyObject *m, *d;
  m=Py_InitModule ("dougTest", dougMethods);
  d=PyModule_GetDict (m);
  //ErrorObject=Py_BuildValue ("s", "dougTest.error");
  ErrorObject=PyString_FromString ("multiarray.error");
  
  PyDict_SetItemString (d, "error", ErrorObject);
  if (PyErr_Occurred ()) {
    Py_FatalError ("cant init module dougTest");
  }
}




-- 
Doug Epps -- Tippett Studio
510 649 9711 (voice)
510 649 9399 (fax)
doug at tippett.com



From thantos at brimstone.mecha  Mon Aug 23 19:45:09 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Mon, 23 Aug 1999 23:45:09 GMT
Subject: [MP3]  Looking for ID3v2 module
Message-ID: <slrn7s3n27.ad8.thantos@brimstone.mecha>

Has anyone already implimented an ID3v2 Object manipulator?  I have
the ID3v1 stuff that's been floating around, but v2 is much, much more
complex.  I'd hate to reinvent the wheel ...

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From schuerig at acm.org  Wed Aug 25 14:32:21 1999
From: schuerig at acm.org (Michael Schuerig)
Date: Wed, 25 Aug 1999 20:32:21 +0200
Subject: UML
References: <37B96A3A.796BB034@compuserve.com> <mmNw3.12$z3.2904@newsfeed.avtel.net>
Message-ID: <1dx401j.1gttza6ybbcq1N@[192.168.1.2]>

Phil Mayes <nospam at bitbucket.com> wrote:

> UML is a standardised set of graphical representations promulgated by the
> gurus who specialise in selling high-priced snake-oil solutions to big
> organisations.  They have simply brought together a number of traditional
> methods such as state diagrams and decreed the detailed format.  To judge
> their skill level, consider this: they chose to draw their class
> inheritance trees with the arrows going up towards the root.

Yeah, and there's a reason for the direction of the arrows.
Generalization and implementation relationships are special cases of
dependencies. And dependencies point from the dependend entity to the
one it depends on -- that's how it usually is.

Michael

-- 
Michael Schuerig
mailto:schuerig at acm.org
http://www.schuerig.de/michael/



From chad at vision.arc.nasa.gov  Thu Aug 19 20:10:16 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Fri, 20 Aug 1999 00:10:16 +0000
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk> <37BC8D05.142E1DB0@aw.sgi.com>
Message-ID: <37BC9CE8.67097FC6@vision.arc.nasa.gov>

Gary Herron wrote:

> Stuart Reynolds wrote:
> >
> > Does anyone know how to tell if Tk has already instaniated a Toplevel
> > window? I want to be able to open several Toplevel windows in any order.
> >
>
> Perhaps the easiest way is to never use the default top level window:
>
> from Tkinter import *
> DefaultTop = Tk()
> DefaultTop.withdraw()  # Never to be seen

Just be careful if you do this, to remember that if you are relying on the
behavior of closing all the visible windows to end your program, it won't.
The withdrawn() window is still running mainloop(), and needs to be
interrupted somehow.  There are some ways around this using callbacks,
etc.  So, depending on the behavior you prefer, this method or my method
each have their strengths.

Chad

PS.  What is the way to register a callback so that you can intercept the
window manager's request to destroy a window?  Then you could just bind
the sys.exit call to this on the Tk object...





From tim_one at email.msn.com  Mon Aug 30 02:25:30 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 30 Aug 1999 02:25:30 -0400
Subject: fo.flush() & fo.write() on windows
In-Reply-To: <37C9DE12.7498090E@rl.gov>
Message-ID: <000401bef2b0$754b7d40$992d153f@tim>

[Tim]
>> file.write() always returns None, as documented.  Don't know what
>> you expected it to return, but read the docs.

[Roberts, Robert J]
> I must have been reading the wrong place in the documentation all
> this time. I was reading doc/lib/node105.html ...

Python offers a rich variety of I/O primitives (how's that for positive
phrasing <wink>?).  You're using the .write() method of file objects, but
reading the docs for the low-level os.write function, which is a different
beast entirely.

Alas, this stuff is spread out.  The docs for the builtin "open" function (and
its optional 3rd buffering argument) are in section 2.3 (Built-in Functions).
The docs for file object methods are in section 2.1.7.9 (File Objects).

>> The only way to get [writes committed to disk] under Windows is to
>> pass the non-standard "c" (for Commit) mode flag to "open":
>>
>>     fo = open(foName, 'ac')
>>
>> [and then they still may not show up on disk until you do fo.flush()]

> I'll look into this too, but I don't want to slow the code down
> anymore if I can help it. I'll compare it with closing and reopening
> the file (which DOES dump the buffer immediately) just to see what
> happens.

Probably depends on whether you need stuff reflected on disk after every single
write, or can afford to bunch up a few writes before flushing.  close+open is
certainly more portable, but more likely slower more often.  OTOH, I don't see
anything in MS's tech docs that *promises* closing a file will immediately
write it to disk; they do promise flush will write to disk immediately if you
use "c" mode.

windows-it's-not-just-an-os-it's-an-adventure-ly y'rs  - tim





From aahz at netcom.com  Thu Aug 19 01:01:59 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Aug 1999 05:01:59 GMT
Subject: Unsorting(randomizing) a sequence
References: <000701bee917$d8161cc0$eaa22299@tim>
Message-ID: <7pg347$pq5@dfw-ixnews13.ix.netcom.com>

In article <000701bee917$d8161cc0$eaa22299 at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>
>I'll save you a near-infinity of work, then <wink>:  Python's Wichmann-Hill
>uniform generator has a period of around 1e12 to 1e13.  Since the point at
>which you begin in the period completely determines the permutation that
>will be generated, you can't get more than 1e13 distinct permutations out of
>it.  So a sequence of length 16 is the longest for which it could even
>conceivably produce all permutations; if you use it to, e.g., shuffle a deck
>of cards, the number of permutations any algorithm based on WH can get back
>is a vanishingly small fraction of the possibilities (since 52! ~= 1e66; we
>can only get back at most 1 of each 1e66/1e13 = 1e53 possibilities).

What happens if you re-seed?

I'm curious because I just wrote some code that re-seeds every hundred
calls to random().  In my case, though, it's necessary because I
actually don't care that much about randomness per se; I'm using it to
generate unique numbers (with duplicate detection), and I need to ensure
that if I ever do happen on the same random number sequence as a
previous run of the process, it gets aborted reasonably quickly.

(This is a temporary hack to deal with an unfortunate choice in hashing
functions until we can switch to some code that uses SQL Server GUIDs.
And, yes, I did find out about Python's hash() function *after* I wrote
this....  <sigh>)
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From tim_one at email.msn.com  Mon Aug 30 20:56:18 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Mon, 30 Aug 1999 20:56:18 -0400
Subject: pop & append
In-Reply-To: <AEjBCBAaxpy3EwFp@jessikat.demon.co.uk>
Message-ID: <000401bef34b$a2df6a00$492d153f@tim>

[Robin Becker]
> Someone from the Zope list tells me that the list pop and append
> operations are thread safe. Is this the case?

All list methods are atomic.

> How can one tell if particular bits of python code are 'atomic'?

Decades of experience <wink>.  Save a few years by reading FAQ entry 4.88
first.

or-just-search-the-faq-for-"atomic"-ly y'rs  - tim





From nosleep at atl.mediaone.net  Tue Aug 31 21:56:05 1999
From: nosleep at atl.mediaone.net (Kristopher Johnson)
Date: Wed, 01 Sep 1999 01:56:05 GMT
Subject: calling python interpreter
References: <37cbccd8.5014784@news.tiscalinet.it> <37ccefcf.1207310@news.tiscalinet.it>
Message-ID: <37CC87C2.75DC8993@atl.mediaone.net>

Kranio wrote:
> 
> I found a solution: chmod ugo+rx pippo.py. But ... should I do the
> same for every script? Does exist a faster system?
> 
> Marco

Write a Python script to do it. :-)

- Kris



From tchrist at mox.perl.com  Fri Aug 20 10:54:50 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 20 Aug 1999 07:54:50 -0700
Subject: carpe linguam
Message-ID: <37bd5e2a@cs.colorado.edu>

Private mail has indicated a remarkably amount of distrust about my
learning Python.  Some verges on the unbelievable.  

And yes, I react to personal digs unquietly.  That this should
come as a surprise is itself a surpise.

The first and smaller thing I wish to say is that when I mentioned
"sifting through the dreck for gems", I meant sorting through the flames
that have occasionally been levelled at me, to skip their heat and
continue on in search of useful information on other people's postings.
I was certainly *not* talking about the language as dreck, as one
person thought I meant.

Secondly, and much more importantly, the reason I am learning Python
is *not* because I wish to "score some quick hits and prove to c.l.p.m
that Perl is best after all" [sic] as one private querent asked, to my
complete and utter horror.

You may not have known this of me, but I am a collector of languages,
and to some extent, of language trivia.  Can you honestly believe that
I studied Spanish, French, German, Italian, Portuguese, or Latin merely
to prove to myself and the world that English was the best language?

Answer for yourself the question of why I learned those languages,
and you'll understand my motivations on learning Python.  Think about
it for a bit, please.

Done thinking already?  Sigh.  Very well, it's quite simple, really:

 *  I want to be able to talk to other people in their own language.
 *  I want to read their literature.  
 *  I want insights into my own tongue.

Is that really so terribly difficult to identify with?  No conspiracy
theories of nefarious purposes are required.

Well, that's it for now.  The plane beckons, and I'm off to Monterey.
I hope to see some of you there.

--tom
-- 
"They always had a rational explanation for the haunting, but they had a dog
 that could talk."
				- My comment about Scoobie-Doo



From davidopp at megsinet.net  Thu Aug 26 10:59:22 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Thu, 26 Aug 1999 10:59:22 -0400
Subject: A Plea For A Return To Sanity!!!
References: <37bc1339@cs.colorado.edu> <37bc313b@cs.colorado.edu> <14268.17944.323020.242210@bitdiddle.cnri.reston.va.us> <019301beeaea$004c4d80$f29b12c2@secret.pythonware.com> <37bd4876@cs.colorado.edu> <MPG.1229cd5eed750a81989687@news.mop.no>
Message-ID: <37C55649.8320449D@megsinet.net>

Dear Group,

How we have let the ravings of a few PERL programmers postings shift the
entire tide of this newsgroup to comparing Python & Perl and throwing various
insults I do not know....Can we stop this now and get back to discussing how
to better program in Python or am I misinterpreting the real reason for the
existence of this group?!


David O.

Alexander Staubo wrote:

> In article <37bd4876 at cs.colorado.edu>, tchrist at mox.perl.com says...
> >      [courtesy cc of this posting mailed to cited author]
> >
> > In comp.lang.python,
> >     "Fredrik Lundh" <fredrik at pythonware.com> writes:
> > :as an aside, it's a bit sad that Tom didn't read the book's
> > :preface, where Mark says, among other things:
> >
> > It's a shame that so many of you should be so unbelievably harsh on me
> > when *I'm* the one who with malice toward none and full of revulsion
> > at the flames causes by *YOUR* community's ignorant attacks on mine,
> > that *I* made the honest effort to learn *YOUR* language, and that I
> > furthermore for your benefit not my own then took the additonal trouble
> > to write-up my experiences and observations gleaned
>
> With all due respect -- not being an active part of this discussion, I
> have observed from afar no flames or unreasonable critique, bar an
> obviously benign "Try reading the docs". Do I sense a whiff of paranoia
> in the air?
>
> Your comments about manpages seem to imply that "man" is the defacto
> standard for documentation; it simply isn't. The quasiphilosophical idea
> that "if it's not in the manpages and not in the Lutz book, it doesn't
> exist" [my phrasing] sounds a bit like the tantrums of a person too
> comfortable in his own land to accept the value of others; "If it
> doesn't have McDonald's, it's not real country."
>
> Your tool isn't everybody else's tool. Personally I don't even know what
> "apropos" is. Even if I made an effort at using it, it doubt it would
> make _my_ workday any easier; I have my own tools, and if I had to use
> man (which isn't part of my toolset), I would nonetheless submit to its
> archaic interface and use it, just as you ought to submit to the
> inflexibilities of the HTML documentation. No need to invoke a fatwah on
> anybody; we live to learn about and deal with our differences.
>
> >while five hundred
> > miles away from the nearest tree or internet link.
>
> Python -- or the binary distribution at any rate -- comes with adequate
> HTML documentation in the box, no tree or Internet link required.
>
> [one pompous declamation of self-importance, two generalizations, three
> incriminations against the community, one incident of improper language,
> and three cynicisms generously deleted]
>
> >
> > --tom
> >
>
> --
> Alexander Staubo             http://www.mop.no/~alex/
> "QED?" said Russell.
> "It's Latin," said Morgan. "It means, 'So there you bastard'."
> --Robert Rankin, _Nostramadus Ate My Hamster_




From mlh at idt.ntnu.no  Wed Aug 11 10:11:40 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 11 Aug 1999 16:11:40 +0200
Subject: Cgi with Python
References: <7op9uu$15a$1@garnet.nbnet.nb.ca>
Message-ID: <y0jd7wuh0f7.fsf@drue.idi.ntnu.no>

"Troy Coburn" <troy at quantumtraining.nb.ca> writes:

> Hi there.
> 
> 
> I am trying to set up some cgi stuff using python...I have just started
> using python and I am running into some proplems.
> 
> When I try to run the python script I get the following error:
> 
> HTTP Error 403
> 403.2 Forbidden: Read Access Forbidden
> 
> This error can be caused if there is no default page available and directory
> browsing has not been enabled for the directory, or if you are trying to
> display an HTML page that resides in a directory marked for Execute or
> Script permissions only.
> 
> Please contact the Web server's administrator if the problem persists.
> 
> 
> Help Please

Well - the error message says it all :)

Just set the read and execute permissions for your script.

$ chmod a+rx script.py

> 
> nester at nbnet.nb.ca
> 
> 
> 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From philh at vision25.demon.co.uk  Sun Aug  8 07:27:00 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 08 Aug 99 11:27:00 GMT
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk> <7ojgv8$2drq@enews1.newsguy.com>
Message-ID: <934111620snz@vision25.demon.co.uk>

In article <7ojgv8$2drq at enews1.newsguy.com>
           p.t at iname.com "Peter Torstenson" writes:
> Phil Hunt <philh at vision25.demon.co.uk> wrote in message
> news:934075710snz at vision25.demon.co.uk...
> > I have two python modules, both of which need to refer to classes
> > defined in the other one. How do I do this?
> >
> > I am currently putting at the start of each module a line of the form:
> >
> >    from theOtherModule import *
> >
> > However, this is not working. My modules -- greatly simplified -- look
> > like this:
> >
> >    #bak.py
> >    from hhh import *
> >    ti = AnHhhClass()
> >    print ti
> >
> > and:
> >
> >    #hhh.py
> >    from bak import *
> >    class AnHhhClass: pass
> >
> > When I run ``python bak.py'', I get the error message
> > ``NameError: AnHhhClass''.
> 
> Referring to something in an other module you have to put the modulename in
> front:
> Like in bak.py: "ti=hhh.AnHnnClass()"

Unfortunately this doesn't work, I get ``NameError: hhh''.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From tim_one at email.msn.com  Sat Aug 28 04:05:34 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 28 Aug 1999 04:05:34 -0400
Subject: A Plea For A Return To Sanity!!!
In-Reply-To: <37C6B987.1662D46F@arlut.utexas.edu>
Message-ID: <000201bef12c$1afa0300$ad2d153f@tim>

[Thomas Bryan, about a Python-for-Perlers FAQ]
> ...
> Is anyone putting this together?  Should it be a FAQ Wizard or what?
> I could set up a FAQ Wizard on my Starship page if others are willing
> to contribute.  I must confess that I haven't used Perl much since
> learning Python, so I'd need a lot of help.

You're making it too hard:  Tom already put a great deal of useful work into
his Python postings, and the confusions a-leaping from every other line <wink>
*are* "the FAQ".  All it still needs are point-by-point brief clarifications
snatched from the depths of the hellish flames that followed.

It doesn't need more info; it needs less noise.

they-also-serve-who-only-post-and-nag<wink>-ly y'rs  - tim





From jstevens at basho.fc.hp.com  Fri Aug 13 23:03:46 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:03:46 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132003.OAA03089@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:03 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ce82@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :With PCRE, I prefer Python.  I am slowly giving up Perl altogether,
> :except as a training language in OO classes.  Not to surprisingly,
> :Python is preferred more than 7 to 1 over Perl by students who are
> :exposed to both at the same time.
> 
> I'm sure I am perfectly capable of presenting both Perl and Python
> in such a way that the students would prefer Perl by 7 to 1 as well.

Except that in your case, it would a concious attempt to influence
the outcome.

In my case, it was not.  In fact, the first four times I did this,
I was subconciously favoring Perl.  Imagine my shock. . .

> So what?

So. . . draw your own conclusions.  I've already seen how you
structured one reply.

John S.




From kernr at mail.ncifcrf.gov  Wed Aug 25 14:37:01 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Wed, 25 Aug 1999 18:37:01 GMT
Subject: problem compiling extension on Win32
References: <7pvja1$u8e$1@nnrp1.deja.com>
Message-ID: <37c43742.14603383@news.erols.com>

On Wed, 25 Aug 1999 02:10:17 GMT, g_v_wilson at my-deja.com wrote:
>I am trying to build a small Python extension
>in C on Windows NT using the Microsoft 'cl'
>compiler from the command line (long story).
>Everything is working, except my attempts to
>#define USE_DL_EXPORT and then use DL_EXPORT.
>I have tried #defining the former both before
>and after #including Python.h, but DL_EXPORT's
>expansion is never correct.  Has anyone run
>into this?  If so, I'd be grateful for help...

Nope.

>Please reply directly to:
>
> g v w i l s o n @ i n t e r l o g . c o m
>
>as I am not a regular reader of this group.
>
>Thanks,
>Greg
>
>p.s. anyone out there building Fortran extensions
>on Windows with G77?

Here, I can help you.  See
http://starship.python.net/crew/kernr/mingw32/Notes.html

E-mail me if you need more specific help.

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From tdixon.no at spam.fwi.com  Thu Aug 26 10:49:28 1999
From: tdixon.no at spam.fwi.com (Timothy Dixon)
Date: Thu, 26 Aug 1999 14:49:28 GMT
Subject: PyKDE for Win ?
References: <7q1als$id9$1@ash.prod.itd.earthlink.net> <37c45fb2.185911967@midpoint> <slrn7s9uuf.q2.moritz@Gertrud.chaos.wg>
Message-ID: <37c553d4.248408843@midpoint>

On Thu, 26 Aug 1999 10:33:19 +0200, moritz at news.uni-mannheim.de
(Moritz Moeller-Herrmann) wrote:

>On Wed, 25 Aug 1999 21:28:13 GMT, Timothy Dixon <tdixon.no at spam.fwi.com> wrote:
>>On 25 Aug 1999 17:55:08 GMT, kikutani at sprintmail.com (Kikutani Makoto)
>>wrote:
>
>>>Looks like a ridiculous qestion, but is there any plan to
>>>port PyKDE to Win world ?
>
>>Um.... Unless I'm wrong (which has happened a nonzero number of
>>times), PyKDE is bindings to the KDE Window Manager in XFree86.  Since
>>KDE doesn't exist in the Windows world, what good would PyKDE do?
>
>KDE is not a window manager. KDE is a desktop environment besed on the QT
>widget set. It works with any X and on most unices.
>

KDE also includes a window manager.




From bowman at montana.com  Sat Aug 28 10:04:20 1999
From: bowman at montana.com (bowman)
Date: Sat, 28 Aug 1999 08:04:20 -0600
Subject: tkinter/python question
References: <7q80gl$t10$1@hermes.louisville.edu>
Message-ID: <37C7EC64.DF009B49@montana.com>

Jeff -=Machtyn=- Gosnell wrote:
> 
> Which type of user is tkinter built for?

In my case, someone who wants to create GUIs using Python. I do not know
Tcl/Tk. When searching for a scripting language, or whatever you wish to
call Python, to supplement my primary language (C), Tcl didn't even make
the short list. My first criteria: does this language resemble C?

I tend to look through Tk and think in terms of the X/Motif type of
operations which underlie it. The syntax is not the greatest, but it
does work after you learn how to use it.

I have looked at wxPython, pyGTK, and other variants. While they are
attractive, I am not willing to devote the time to learning yet another
windowing API, which may or may not make the final cut. Also, Tcl/Tk
tends to be installed on many machines, while GTK+ and wxPython doesn't.
Further, I have a personal tendency towards simplicity, and prefer a GUI
that gets the job done using simple widgets. 

That is one take on Tkinter. I am sure there are many more.



From tim_one at email.msn.com  Fri Aug 27 00:06:22 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 27 Aug 1999 00:06:22 -0400
Subject: Library philosophy
In-Reply-To: <37C57A80.D5C62704@prescod.net>
Message-ID: <000101bef041$85eb8da0$592d153f@tim>

[Paul Prescod]
> ...
> I believe that Python's libraries are in many places too thin of
> a layer on top of the C library and are thus not very Pythonish.
> Evidence of this includes indecipherable function names like
> strtfmt and APIs that return tuples instead of objects and take
> integers as "flags" instead of strings or objects.
>
> Would it it make sense to start wrapping the low-level modules
> with high-level, easier to use modules that would be "the default"
> in Python 2?

"Yes" from me, but it requires a major commitment to API design,
implementation, and-- last but most important --docs.  The great advantages of
sticking to thin layers at the start were that the Unix/POSIX docs were
incorporated by implicit reference, and failures could be blamed on the
platform libc ("hey, don't look at us -- it's just a thin wrapper" <wink>).

roll-your-own-api-and-the-docs-must-come-first-ly y'rs  - tim





From philh at vision25.demon.co.uk  Fri Aug 20 14:09:12 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Fri, 20 Aug 99 18:09:12 GMT
Subject: Book -- or website -- about Tk & tkinter
Message-ID: <935172552snz@vision25.demon.co.uk>

Can anyone recommend a good book or website that explains how Tk does 
its geometry management (the grid, pack etc managers)?

I'd be particularly interested in anything that talks in terms of 
tkinter.

-- 
Phil Hunt....philh at vision25.demon.co.uk




From arcege at shore.net  Sat Aug 21 08:35:05 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Sat, 21 Aug 1999 12:35:05 GMT
Subject: What is a troll by the way? [offtopic]
References: <zelv3.13444$x04.816752@typ11.nn.bcandid.com> <37bdeb1b.31915768@news.erols.com> <37BDF3A7.75FEA04F@vision.arc.nasa.gov>
Message-ID: <Z1xv3.11$EN1.7779@news.shore.net>

Chad Netzer <chad at vision.arc.nasa.gov> wrote:
: Robert Kern wrote:

:> _Troll_ is a USENET term for someone that deliberately posts an
:> inflammatory message simply to stir up argument.

: I've found it funny to watch this term evolve over the past few years.
: It started (I believe) as a reference to the fishing term, and used to
: mean "trolling for flames" (ie. hoping to bring flames to the
: discussion), perhaps employing "flame bait" for this purpose.

Anyone remember BIFF?

: It is now used by many (I believe), as a derogatory reference to someone
: being like an ugly, flesh-eating beast that lives under a bridge. :-)
: And that calling someone a "troll" is simply meant to imply they are
: contemptible for trying to incite flame wars (much like a Troll is
: contemptible for, well, being a Troll).

It is the same with me; I was around during the Spam Wars, when the
term "spam" was first created (yes, it was after the Monty Python
sketch).  My mother never served up spam when I was a kid, so I
couldn't really participate that much, but it was fun to watch (I
wasn't an usenet admin until 2-3 years later).  Originally, from the
history that I remember, some UC-Berkeley students were up late
drinking and watching Monty Python.  They were drunk and decided to
post some "silly spam recipes", crossposting to a bunch of (unrelated)
newsgroups.  Sending silly spam recipes became popular, then turned to
real recipes, but always crossposted to more and more newsgroups.  This
was circa 1986-87, usenet was largely 1200 baud modem based (except the
backbones) and available disk space was hardly what it is now.  Usenet
admins were throwing fits and even started closing down their servers
(if they weren't closed already from the flood of duplicate postings).
It wasn't until a privateer sent out a commercial posting (which was
"illegal" on usenet at the time) to a large number of newsgroups did
the term "spam" really get a negative connotation.

: Anyway, I personally like the first usage, and sometimes lament that
: it is evolving toward the second.  I used the term "troller", rather than
: "troll", but it is probably a lost cause. :-)

Like the difference between "hacker" and "cracker" sometimes. :(

  -Arcege




From kholwerd at xs4all.nl  Mon Aug  9 19:34:52 1999
From: kholwerd at xs4all.nl (kholwerda)
Date: Mon, 09 Aug 1999 23:34:52 +0000
Subject: ANNOUNCE:  wxPython 2.1b2
References: <024c01bee2a0$ec84bf40$1a25d2d1@jenkondev.com>
Message-ID: <37AF659C.BB2444C5@xs4all.nl>

>  or can be used in in
> situations where Python is embedded in a wxWindows C++ application as an
> internal scripting or macro language. Currently supported GUIs are Win32 and
> GTK/X-Windows.

Is there any documentation one how to do this, i have not found it in the python
distribution recently.
I think i need some help here.

Anyway the way wxPython is integrated with wxwindows lookes great, thanks!

I hope it will have a great future!

Klaas





From aa8vb at yahoo.com  Wed Aug 25 14:42:43 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Wed, 25 Aug 1999 14:42:43 -0400
Subject: Which GUI-toolkit
In-Reply-To: <37C4066E.9EB52B39@mpi-dortmund.mpg.de>; from Marc Saric on Wed, Aug 25, 1999 at 05:06:22PM +0200
References: <37C4066E.9EB52B39@mpi-dortmund.mpg.de>
Message-ID: <19990825144243.A137012@vislab.epa.gov>

Marc Saric:
 |Just a newbie question: I want to write a GUI-tool for a
 |FORTRAN-program-package (communication goes via input-and configuration
 |files for that program for now) and I had a look at wxWindows and
 |Tkinter. 
 |To me it seems that wxWindows provides a lot more prebuild highlevel
 |parts like notebooks and tree views. On the other hand Tk might be more
 |widespread, especially on UNIX (I plan to do the program for both UNIX
 |and Windows with a focus on UNIX due to the FORTRAN package).
 |
 |Any advice to which GUI I should use??

If Tkinter and wxWindows are your candidates, I'd recommend reading the
Tkinter vs. wxPython thread from June:

     http://x47.deja.com/=dnc/dnquery.xp?ST=PS&QRY=*&defaultOp=AND&DBS=1&format=threaded&showsort=score&maxhits=100&LNG=ALL&subjects=tkinter+%26+wxpython&groups=*python*&authors=&fromdate=&todate=

A little long, but averaging it all together you'll probably get a
well-rounded perspective.

Randall



From gmcm at hypernet.com  Sun Aug 22 19:46:17 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 22 Aug 1999 18:46:17 -0500
Subject: Date Calculations
In-Reply-To: <37BF5D1A.9865CDF@fnmail.com>
Message-ID: <1276760785-12130868@hypernet.com>

Per Kistler wrote:
> 
> I'm making some Date Calculaton class with things like
> deltaDays, addDays, daysInYear and so on, above the calendar 
> and time classes. After some 300 lines I thought, I better
> ask, whether that has already been done by someone else.
> How is this?

Well Jeff Bauer just talked about his date classes in the last couple 
of days, and Marc-Andre Lemburg has a C extension that does anything 
you could possibly think of. As I recall, they both have their stuff 
on their homepages at starship.python.net.

- Gordon



From provost at VirtualPrototypes.ca  Wed Aug 25 16:48:06 1999
From: provost at VirtualPrototypes.ca (Paul Provost)
Date: Wed, 25 Aug 1999 20:48:06 GMT
Subject: [MP3]  Looking for ID3v2 module
References: <slrn7s3n27.ad8.thantos@brimstone.mecha>
Message-ID: <aEYw3.3673$p44.130407@carnaval.risq.qc.ca>

There's a library available at http://www.id3v2.org . Anyone want to wrap it
into an extension?

Paul

Alexander Williams <thantos at brimstone.mecha> wrote in message
news:slrn7s3n27.ad8.thantos at brimstone.mecha...
> Has anyone already implimented an ID3v2 Object manipulator?  I have
> the ID3v1 stuff that's been floating around, but v2 is much, much more
> complex.  I'd hate to reinvent the wheel ...
>
> --
> Alexander Williams (thantos at gw.total-web.net)
> "In the end ... Oblivion Always Wins."





From m.faassen at vet.uu.nl  Sat Aug 28 14:03:33 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 28 Aug 1999 18:03:33 GMT
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net>
Message-ID: <7q989l$7bi$4@newshost.accu.uu.nl>

Moshe Zadka <moshez at server.python.net> wrote:
> On Sat, 28 Aug 1999, Tim Peters wrote:

>> python-folk-are-the-best-buddies-psu-dues-can-buy-ly y'rs  - tim
>                                    ^^^
> There are a few options:

> 1. The mailman at Python org had a glitch

That's Python software, right? Python software can't have glitches. Nonsense.

> 2. The timbot made a type

The timbot making a typo? A bot? Nonsense.

> 3. There's this underground association, Python Software Union, which goes
> around breaking thumbs of people who use Python without paying the PSA.

Then this is left.

Uh oh..Where is that form..

Regards,

Martijn




From m.faassen at vet.uu.nl  Sat Aug 14 19:15:59 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 14 Aug 1999 23:15:59 GMT
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <37b5a8e0@cs.colorado.edu>
Message-ID: <7p4tbf$kre$1@newshost.accu.uu.nl>

Tom Christiansen <tchrist at mox.perl.com> wrote:
>      [courtesy cc of this posting mailed to cited author]

> In comp.lang.python, Drew McDowell <drew.mcdowell at msfc.nasa.gov> writes:
> :I looked in the FAQ and couldn't find anything on the Compound Assignment
> :Operators.  (+=, *=, -=, etc..)  Why dosen't Python support them?
> :Is there a easy way to add them?

[motivations against assignment in expressions snipped]

There's an interesting analogy here with the inability to leave off the
{ and } in Perl in a single lined block, like you can in C. The motivation
is that people shoot themselves in the foot with it too often, I recall.

Note that the usual way to do such a readline loop in Python goes like
this:

while 1:
    line = f.readline()
    if not line:
        break
    ... stuff with line here ..

Not too difficult and easy enough to recognize, though it does take some
getting used to.

There were a couple of big threads about assignment in expressions
a while back. In one of those discussions people came up with
interesting ways to do these things without the infinite loop. Often
these center on iterator approaches. It's possible the vaporware Python 2
will extend on that idea.

> As for why you can't write the statement (note: not the expression)

>     x["every"]["good"]["boy"]["does"]["fine"] =  (
> 	x["every"]["good"]["boy"]["does"]["fine"] + 1 )

> as 

>     x["every"]["good"]["boy"]["does"]["fine"] += 1

> The basic reason seems to be that there are some people who believe that
> this kind of thing is cruel and unusual punishment levied upon the poor
> Pascal(ish) programmers purely for sake of brevity.

You can do:

a = x["every"]["good"]["boy"]["does"]
a["fine"] = a["fine"] + 1

Not perfect, perhaps, but a lot easier. It's interesting, but I actually
don't recall running into this problem a lot with Python at all.

> Common arguments in support of such ops include:

>     1) This requires care to avoid mis-duplication:
> 	x["every"]["good"]["boy"]["does"]["fine"] =  (
> 	    x["every"]["good"]["boy"]["deos"]["fine"] + 1 )

You avoid misduplication with the above solution.

>     2) Side effects must be duplicated:
> 	x[every()][good()][boy()][does()][fine()] =  (
> 	    x[every()][good()][boy()][does()][fine()] + 1 )

Partially solved by the above solution.

>     3) It's too long to type.

I'd say solved by the above solution.

>     4) Needless redundancy obscures the real meaning.

I seem to be neutral on this. :)

> Whereas common rebuttals include:

>     1) Toss /bin/ed and get yourself a real editor.

Indeed helpful.

>     2) Don't use side-effects, you wicked creature, you!

When you use side-effects, be careful. :)

>     3) It's more efficient to keep a tmpobj anyway.

I don't get this one. By 'tmpobj', do you mean my solution? Indeed,
it'd be more efficient as you'd avoid (costly) Python name lookups.

>     4) Redundancy can clarify through redundancy.

I plead neutrality once again.

> As to why there's no x++ there, consider how often people from a
> Pascal(ish) background are confused between x+1 and x++ and ++x, and
> you'll begin to see what kind of problems caused it to be vetoed.

I don't think Python was designed for people with a Pascallish background
at all, by the way. 

> But the best reason is probably the one another poster already supplied:
> because Guido didn't like them.  An interesting study would be to check
> the extent to which he also forbids such constructs in his own C code. :-)

When coding in C, code in C; when coding in Python, code in Python. From
what I've heard the Python C sourcecode is well designed; I'm also sure the 
constructs are used in many places.

But again, this problem is something that in practice just does not seem
to happen to me when I program in Python.

Regards,

Martijn




From mark.mcmahon at eur.autodesk.com  Tue Aug  3 07:32:51 1999
From: mark.mcmahon at eur.autodesk.com (mark.mcmahon at eur.autodesk.com)
Date: Tue, 3 Aug 1999 13:32:51 +0200 
Subject: Getting "Short" filenames inWindows
Message-ID: <C2AA61E5CAA7D211832F0008C7A4138E01B4873C@eurmsgneu01.eur.autodesk.com>

Hi Everyone,

Is there some simple way to get the short path name
from the long one in Windows.

os.path, dospath and ntpath files do not seem to
allow one to access the short name easily.

Thanks
	Mark



From stephan at pcrm.win.tue.nl  Mon Aug 30 03:55:00 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 30 Aug 1999 09:55:00 +0200
Subject: Question: lists in classes
References: <37CA1803.2FA9261B@ug.cs.su.oz.au>
Message-ID: <m3k8qdhfe3.fsf@pcrm.win.tue.nl>

Jeff Turner <jturner at ug.cs.su.oz.au> writes:

> Apologies if this has been answered before...
> 
> class foo:
>     list=[]
> 
> a=foo()
> b=foo()
> a.list.append(123)
> print b.list
> 
> I would expect the above code to return "[]", since b has not been
> modified since instantiation. However it returns [123].

Sure, `b' has not been modified. However, a.list and b.list are
both references to the same list object. You changed the object
through a.list, so now b.list reflects the change as well.
 
> How would I tell python that I only want to _define_ the class, not
> instantiate it?

Everything you put directly in the class body is executed when the
class is defined. If you want to perform something when an instance
is created, put it in the class' __init__ method.

The following gives the behaviour you probably expect:

class foo:
    def __init__(self):
        self.list = []

Now a.list and b.list are distinct objects.

Greetings,

Stephan




From den at analyt.chem.msu.ru  Mon Aug 23 05:50:27 1999
From: den at analyt.chem.msu.ru (Denis S. Otkidach)
Date: Mon, 23 Aug 1999 12:50:27 +0300
Subject: locale on NT
Message-ID: <7pr5h3$cgc$1@news.msu.ru>

Why locale module on Windows NT seems to be broken? The registry entries
contain all necessary information...





From xah at weborder.com  Sun Aug 29 17:10:14 1999
From: xah at weborder.com (Xah Lee)
Date: Sun, 29 Aug 1999 14:10:14 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu>
			<7p49l5$uv7$1@nnrp1.deja.com>
			<37b634a2@cs.colorado.edu>
			<FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net>
			<37C552DF.9F90BA36@megsinet.net>
			<slrn7sblvo.p96.wtanksle@dolphin.openprojects.net> <14277.63776.733073.298192@dolphin.mojam.com> <37C60262.C6771B75@megsinet.net>
Message-ID: <37C9A1B6.1327BCCD@weborder.com>


David O. wrote:
> 
> Where can I find info on Minotaur?
> 
>David O.

Dear David O'moron,

You can find Minotaur info at this url:
 http://www.pantheon.org/mythica/articles/m/minotaur.html

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html



From jstevens at basho.fc.hp.com  Fri Aug 13 23:08:17 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:08:17 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132024.OAA03817@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:24 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cf91@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :Due to this feature alone, Python programs are easier to read by
> :non-authors than Perl programs are.
> 
> Prove that.  Stop asserting it.  Prove it.
> 
> Here's your challenge.  Pick a Perl programmer with experience equivalent
> to mine, and show me their code.

Of course, there are very few Perl programmers that *HAVE* experience
equivalent to yours.

So doing what you suggest would prove nothing.

Perform the same test across even a randomly-selected sample, and
you will begin to see the pattern.

> I'm sure I'll find it perfectly legible.

Even for you, I doubt that.  One Perl author in one of my environments
used no "unnecesary" white space at all.  None.  His code was
incomprehensible, even to the guy we brought in to teach Perl.

> I say this for certain because I *do* read their code, and it's completely
> easy to do so.

Yeah.  Right. ;->

Been there, gathered the statistics.

> If you're talking about someone who can't stop treating Perl like
> BASIC or FORTRAN, and just doesn't `get' it, then sure, it's unnatural.
> But that's hardly my fault that they don't know Perl, now is it?

The fact that Perl lets them write unmaintainable code *IS* your
fault, when state of the art in language design can and does prevent
some of this.

What, you've never heard of a C coding style?  Never seen one enforced?

Never, in your life, compared productivity levels, defect rates, and
repair costs across C or Perl code that has been written to an enforced
coding standard, vs. code that was totally free form?

Perhaps you should.

I have.

John S.




From philh at vision25.demon.co.uk  Tue Aug 31 18:52:47 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 31 Aug 99 22:52:47 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7ps4ma$n06$1@wagner.wagner.home> <935457373snz@vision25.demon.co.uk> <gr6gq7.jd.ln@bigbird.sw.asiainfo.com>
Message-ID: <7qmpf1$f9$2@newnews.ilk.de>

In article <gr6gq7.jd.ln at bigbird.sw.asiainfo.com>
           markh at usai.asiainfo.com  writes:
> Phil Hunt <philh at vision25.demon.co.uk> wrote:
> : In article <7ps4ma$n06$1 at wagner.wagner.home>
> :            vitus at wagner.rinet.ru "Victor Wagner" writes:
> 
> :> Really it is very simple to define parrot interpreter on tcl - just
> :> define commands such as window, columnLayout, rowLayout as tcl procedures 
> :> which substitute
> :> something by default.
> 
> : Would anyone want to do this in Tcl as opposed to Python? (Not
> : wishing to get into another language war :-)).
> 
> Interestingly, your parrot specifications follow Tcl's syntax,

Great minds think alike, eh? :-)

BTW, I might have a bare bones implementation of the tkinter
backend up by this weekend, if all goes well.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From fredrik at pythonware.com  Thu Aug 26 06:59:36 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 26 Aug 1999 10:59:36 GMT
Subject: [Newbie]:Need help with "slicing"
References: <7q34u6$fd4$1@nnrp1.deja.com>
Message-ID: <00ef01beefb2$16e16540$f29b12c2@secret.pythonware.com>

<gtnorton at my-deja.com> wrote:
>         Trying to figure out how to slice lists.  For example:
> 
>           >>> L = ['help', 'with', 'slice']
> 
>                        How do I slice to just show:
> 
>           >>>['help', 'slice']1st and last -with L[?:?]

a slice is everything *between* two indices.

to get the result you're after, you can either
*remove* a slice from your list:

del L[1:-1]
print L

or concatenate two slices (cheaper than you
might expect, since all Python's doing is copying
references around):

print L[:1] + L[-1:]

</F>




From kas at maps.magnetic-ink.dk  Fri Aug  6 08:51:09 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Fri, 6 Aug 1999 14:51:09 +0200
Subject: Help with dictionary, please
References: <7oek07$so3$1@alexander.INS.CWRU.Edu>
Message-ID: <slrn7qlmht.35c.kas@titan.magnetic-ink.dk>

Rob Lake writes:

> 	 	   key = (prev2,prev1)
> 	           dict[key] = dict[key] + 1
>    and got this error:
> 	  File "Metro:Python 1.5.1:learn2.py", line 34
>           dict[key] = dict[key] + 1
>                          ^
> 	SyntaxError: invalid token

The variable key should be a string, you're supplying a tupple.


  //Klaus

-- 
???[ Magnetic Ink ]????????????????????????????????????????????????? ><> ???
???[ http://www.magnetic-ink.dk/download/ ]?????????????????????????????????



From fdrake at acm.org  Thu Aug 19 17:57:06 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 19 Aug 1999 21:57:06 GMT
Subject: More random python observations from a perl programmer
In-Reply-To: <37bc71cc.19700984@news.erols.com>
References: <199908191819.MAA04074@jhereg.perl.com>
	<37bc71cc.19700984@news.erols.com>
Message-ID: <14268.32178.380210.233387@weyr.cnri.reston.va.us>

Robert Kern writes:
 > Manual.  Unfortunately, what it doesn't mention is the URL for getting
 > this documentation.  This is a holdover from the time before the
 > source and documentation were split into separate distributions.  If
 > nothing else is done about manpages, this oversight should be fixed.

Robert,
  Thanks for pointing this out; I'll certainly add the missing
information to the man page.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From gmcm at hypernet.com  Fri Aug  6 16:36:17 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Fri, 6 Aug 1999 15:36:17 -0500
Subject: __init__ question
In-Reply-To: <000101bee041$d5d51440$43f9143e@pc58-mikael.fas.flygt.com>
Message-ID: <1278154513-26981346@hypernet.com>

Mikael Aronsson writes:
> 
> I have a small Python class called Vector, it is a simple 3D vector
> class with an x,y and z variable.
> 
> I would like to have multiple constructors for this (like you can in
> C++).
> 
> Example:
> 
> a = Vector()  # x=0 y=0 z=0
> a = Vector( 1)  # x=1 y=1 z=1
> a = Vector( 1, 2, 3)  # x=1 y=2 z=3
> 
> Can this be done in Python, or can I only have one __init__ method 
> ?

Just one, but 
 
 def __init__(self, x=0, y=0, z=0):

should get you most of the way there. Before you go wild with default 
arguments, read the FAQ on the dangers associated with using a 
mutable variable as a default argument, (not a problem here, of 
course).




- Gordon



From beall at psych.ucsb.edu  Wed Aug  4 12:32:09 1999
From: beall at psych.ucsb.edu (Andy Beall)
Date: Wed, 04 Aug 1999 09:32:09 -0700
Subject: building embedded classes as in C/C++
References: <37A78579.9856D16@psych.ucsb.edu> <al8907sqh81.fsf@sirppi.helsinki.fi>
Message-ID: <37A86B09.CB163857@psych.ucsb.edu>

Yes, thanks for the reply--I now see what I was doing wrong.  What
through me off (besides not reading thoroughly enough) was that non-list
and non-class variables are not class wide when declared using the
convention I did.  I understand now how to do what I want but why do
simple scalar-like variables work one way and lists and classes
another.  I'm definitely missing some subtlety of Python here...

For example,

class Object:
       p = 0 
       pos = Point()
       ang = Point()

Here, 'p' does not seem to be class-wide while 'pos' and 'ang' are.  Why
the discrepancy?

Thanks for any enlightenment,
Andy-



Markus Stenberg wrote:
> 
> Andy Beall <beall at psych.ucsb.edu> writes:
> > I'd like to build an embedded structure (Python class) much like I do in
> > C if possible.  When I do the following, multiple instances of my final
> > class overwrite each others sub-class values!
> 
> You did not pay attention, however; you set _class_wide_ variables, global
> to all instances (pos, ang). Thus, the behavior was to be expected.
> 
> Try this instead:
> 
> class Object:
>         def __init__(self):
>                 self.pos = Point()
>                 self.ang = Point()
> 
> >



From moun at usenix.org  Wed Aug 11 12:53:38 1999
From: moun at usenix.org (Moun Chau)
Date: Wed, 11 Aug 1999 16:53:38 GMT
Subject: 2nd Conference on Domain-Specific Languages, October 3-5, 1999
Message-ID: <FGB89E.24z@usenix.org>

*Take part in the community dedicated to promoting the promise
  and practice of domain-specific languages
*Learn of current advances of DSLs and examine the spectrum of
  benefits which DSLs can provide
*Concentrate on the unique aspects of DSL design, DSL
  implementation, and of DSLs as software engineering tools

DSL '99
2ND CONFERENCE ON DOMAIN SPECIFIC LANGUAGES
October 3-5, 1999
Omni Hotel, Austin, Texas, USA
-----------------------------------------------------
Save.  Register by September 13, 1999.
Register on line:  www.usenix.org/events/dsl99
-----------------------------------------------------
Sponsored by USENIX, The Advanced Computing Systems Association
In cooperation with ACM SIGPLAN and ACM SIGSOFT

Refereed Papers Examing the Unique Positive Aspects of DSLs in areas
such as:
* Different Approaches To DSL Construction
* New DSLs For Problem Domains such as Specifying Hardware Circuits
and Robot Control Protocols
* Creating Data-Intensive Web Sites and Collaborative Applications

Keynote:  Towards More Natural Domain-Specific Languages
by Brad A. Myers
Human Computer Interaction Institute, Carnegie Mellon University

Hot Research Reviews
* Domain-Specific Languages for Programming and Security in Active
Networks, by Carl A. Gunter, University of Pennsylvania
* A Methodology for Designing Domain-Specific Languages Using Program
Specialization, by Charles Consel, IRISA/University of Rennes

Insight and Perspective from Invited Talks
* Language Technology for Performance and Security
* The Next 700 Markup Languages

Interactive Sessions, Hosted Luncheon, and Conference Reception

"(Regarding the1997 DSL Conference) Papers presented were of such
uniformly high quality that the program committee was unable to
single out one for special commendation. Interest in the BOFs was
so great that initial plans to run them concurrently were shelved
in favor of a consecutive schedule.  Overall, I believe the 
conference was a great success!"
Andrew J. Forrest, AT&T Labs



From conversy at lri.fr  Mon Aug 30 09:15:07 1999
From: conversy at lri.fr (Stephane Conversy)
Date: Mon, 30 Aug 1999 15:15:07 +0200
Subject: swigging polymorphic collections...
References: <51C9F3C392D0D111A15600A0C99E64A9AF54E0@localhost>
Message-ID: <37CA83DB.476BD350@lri.fr>

very very bad trick (better solution welcome !!) :

what I did:

class SubNode1;

       class Node {
            public:
                int n;
                enum node_type {
                    SubNode1_t
                };
                node_type type;

                SubNode1*
                toSubNode1() {
                    return dynamic_cast<SubNode1*>(this);
                }
            };

        class SubNode1 : public Node {
            public: int sn1;
                SubNode1() : type(SubNode1_t) {}
        };

test the type in python and cast:

    for i in list:
            if(i.type==SubNode1_t):
                j = i.toSubNode1()





Bjorn Pettersen wrote:

> I've got a c++ class hierarchy rooted at Node:
>
>         class Node { public: int n; };
>         class SubNode1 : public Node { public: int sn1; };
>
> and a list:
>
>         class List {
>                 std::vector<Node*> lst;
>         public:
>                 void  append(Node* n);
>                 Node* __getitem__(int key);
>         };
>
> So far everything is fine, however, when my c++ module returns a List, I can
> never access "sn1" even though the actual Node is a SubNode1.
>
> Has anyone found a solution to this?  Should I be using a PyList object
> instead? -- how would I insert the nodes so they would be Python shadow
> classes?
>
> -- bjorn

--
St?phane Conversy
http://www-ihm.lri.fr/~conversy/
mailto:conversy at lri.fr






From vartan at leland.Stanford.EDU  Tue Aug 24 19:29:02 1999
From: vartan at leland.Stanford.EDU (Ryan Arthur Vartanian)
Date: Tue, 24 Aug 1999 16:29:02 -0700
Subject: FDL Fortran Wrapper
Message-ID: <37C32ABE.41C6@leland.Stanford.EDU>

I've heard of an app called FDL which is a fortran wrapper much like
swig for C/C++.  Unfortrunately I have been unable to locate it.  If
anyone knows of the existence or a place I could find it, it would be
greatly appreciated.

Ryan

 
-----------------------------------------------------------

     "I don't mind Microsoft making money, I mind them 
         having a bad operating system."
                              -Linus Torvalds June 13, 1999

Ryan Vartanian
Aeronautics/Astronautics Graduate Student
vartan at leland.stanford.edu
------------------------------------------------------------



From salzr at certco.com  Fri Aug  6 16:35:17 1999
From: salzr at certco.com (Rich Salz)
Date: Fri, 6 Aug 1999 16:35:17 -0400
Subject: embedding
References: <7nstgb$st3$1@nnrp1.deja.com> <lxCq3.706$EG4.94343@news.shore.net>
Message-ID: <rqmhp15hkur75@corp.supernews.com>

The following (excerpted, no error-checking)
 PyObject* mod = PyImport_AddModule("__main__");
 PyObject* mdict = PyModule_GetDict(mod);
 PyObject* obj = get_a_python_object();
 PyDict_SetItemString(mdict, "YOUR NAME HERE", pyd);

This seems simpler than Reilly's method.  So is my code wrong, or (gasp) is
there More Than One Way To Do It?
    /r$






From schuerig at acm.org  Wed Aug 25 14:32:19 1999
From: schuerig at acm.org (Michael Schuerig)
Date: Wed, 25 Aug 1999 20:32:19 +0200
Subject: UML
References: <3F8BA119B48AD11197470008C7A493CB98B502@HQOBGEX1> <37c2afa8.67206110@news.compuserve.com> <slrn7s69p4.jb7.wtanksle@dolphin.openprojects.net> <37C3D89C.D57E9956@abg1.siemens.de>
Message-ID: <1dx3zt1.1r2ipo9b4kt16N@[192.168.1.2]>

Volker Hetzer <volker.hetzer at abg1.siemens.de> wrote:

[Regarding UML books]
> I started with "Objectys, Components and Frameworks with UML".
> Was ok.

[by Alan Wills and Desmond D'Souza, Addison-Wesley 1998]

I'd recommend against this book _for using UML_. It's an excellent book
nonetheless, but it uses UML notation in a non-standard way.

To anyone who's looking for a short overview I'd recommend Martin
Fowler's "UML Distilled". If you want the real thing, though, go for
Booch/Rumbaugh/Jacobson "The UML Language Users Guide"; there's really
no way around this one.

Michael

-- 
Michael Schuerig
mailto:schuerig at acm.org
http://www.schuerig.de/michael/



From g.gruber at xsoft.co.at  Thu Aug 26 05:49:33 1999
From: g.gruber at xsoft.co.at (Gerhard W. Gruber)
Date: Thu, 26 Aug 1999 11:49:33 +0200
Subject: emacs with python mode
Message-ID: <37C50DAD.840B67C5@xsoft.co.at>

I'm using Xemacs 21.0 Beta (August 1998) on Win32 and it works quite
fine (is there a newer version out there?). Now my real question. When
I'm using Python emacs recognices the mode but it doesn't automatically
activate syntax highlighting, even though I activated it in the
preferences (same for C/C++ mode as well). Also I was wondering what to
do about the tabs. When I read in a file from the disk emacs showed it
in proper alignment. When I add new statements emacs automatically
aligns it but it doesn't care about blanks. So it looks on the screen as
if the statments are properly aligned but in truth it is not and python
complains about invalid indention. i.e. One line contains 4 blanks and
the next contains 1 tab instead. Since I adjusted tabwidth to 4 blanks
it looks ok, but it is not. I always have to look into another editor to
see if the indention is correct. How can I force emacs to expand all
leading blanks to the according number of tabs or vice versa. I don't
care wether I have tabs or blanks but I need to have the proper
indention, as you probably know this is a required part of the python
syntax.

Thanks.
-- 
bye,
   Gerhard
email: sparhawk at eunet.at
       g.gruber at xsoft.co.at



From pduffin at mailserver.hursley.ibm.com  Fri Aug 13 04:30:12 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Fri, 13 Aug 1999 09:30:12 +0100
Subject: Anyone else find Tk akward?
References: <37B34DBB.E5A79760@sprint.ca>
Message-ID: <37B3D794.6B53@mailserver.hursley.ibm.com>

Orlando Vazquez wrote:
> 
> Is it just me, or is the way Tk organizes things on the window someone
> weird? I'm just starting to play around with it, and it has me puzzled.
> Wouldnt it be a lot easier to just use a coordinte system?
> For example put a button at (0,0) with length 10 and width 20 etc..
> Are there any reasons why it _has_ to be this way?
> 

When designing a GUI (on something like VB which at least when I was
using it had absolute positions/sizes) I find that 99% of the time is 
getting the positions pixel perfect. Once you have done this you then
either need to stop the user resizing it which is not nice, or you have
to do some VB hackery to get it behaving correctly when resizing and it
never works how you expect it to.

Using geometry managers makes life so much easier because they handle
all of the resizing calculations, the difficult part is learning how
to tell the geometry managers what you want to do.

If you absolutely have to have total control Tk (not sure about Tkinter)
has a place manager which lets you put the windows wherever you want to.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From tchrist at mox.perl.com  Fri Aug 13 21:36:17 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 13 Aug 1999 18:36:17 -0700
Subject: Why use Python when we've got Perl?
Message-ID: <37b4ba01@cs.colorado.edu>

Unbelievably stupid subject, eh?  I certainly thought so when it was
shoved at us.  The flame war begun in comp.lang.perl.misc isn't doing
anyone any good, and then coming over to comp.lang.python to stir up the
shock troops to send in a few commandos is even worse.  I don't think
you'd like it if we in comp.lang.perl.misc sent 100,000 script kiddies
over to comp.lang.python to rant at you guys about how K31L they though
Perl was and how L at M3 they thought Python was.  

Do you want to learn or do you want to flame?  Learn?  Really?  Ok, here's
all you have to do.  Go write a medium-sized program comprising many
hundreds of lines, broken up into several files and selectively imported,
but do so in whichever language you know the *least*.  That's the way
to learn, not though wars.  There's no way we can learn from one another
by flaming.  And there's plenty to learn.  Everyone should do this.

Please and thank you.

--tom
-- 
Anyone can be taught to sculpt.
    Michaelangleo would have had to have been taught how not to.
	The same is true of great programmers.
"36 percent of the American Public believes that boiling radioactive milk



From rainer2207 at my-deja.com  Sun Aug 29 05:57:46 1999
From: rainer2207 at my-deja.com (rainer2207 at my-deja.com)
Date: Sun, 29 Aug 1999 09:57:46 GMT
Subject: Function calling problem
Message-ID: <7qb06p$mj5$1@nnrp1.deja.com>

Can someone please explain what's wrong with this code? I've tried to
call the function quit(self) but I get an error. What's wrong here?

from Tkinter import *
import sys, Dialog

class Interface:


	def __init__(self, master):

		self.menu = Menu(master, tearoff=0, name="menu")
		master.configure(menu=self.menu)

		self.frame = Frame(master, width="15c", height="10c",
bd=2, relief="sunken")
		self.frame.pack(fill="both", expand=1)

		m = Menu(self.menu, tearoff=0)
		self.menu.add_cascade(label="File", menu=m, underline=0)

		n = Menu(m, tearoff=0)
		m.add_cascade(label="New", menu=n, underline=0)
		n.add_command(label="RDF Schema", underline=4)
		n.add_command(label="RDF Metadata", underline=4)

		o = Menu(m, tearoff=0)
		m.add_cascade(label="Open", menu=n, underline=0)
		o.add_command(label="RDF Schema", underline=5)
		o.add_command(label="RDF Metadata", underline=5,
command=self.open)

		m.add_command(label="Save", underline=0)
		m.add_command(label="Save As...", underline=5)
		m.add_command(label="Close", underline=0)
		m.add_separator()
		m.add_command(label="Exit", underline=0,
command=self.quit)

		m = Menu(m, tearoff=0)
		self.menu.add_cascade(label="Edit", menu=m, underline=0)
		m.add_command(label="Cut", underline=2)
		m.add_command(label="Copy", underline=0)
		m.add_command(label="Paste", underline=0)

		m = Menu(m, tearoff=0)
		self.menu.add_cascade(label="Help", menu=m, underline=0)
		a = Menu(m, tearoff=0)
		m.add_command(label="About", underline=0)

	def quit(self):

		sys.exit(0)


root = Tk()
inter = Interface(root)
root.title("Title")
root.mainloop()


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From fredrik at pythonware.com  Fri Aug 20 11:33:34 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 20 Aug 1999 15:33:34 GMT
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk> <37BC8D05.142E1DB0@aw.sgi.com> <37BC9CE8.67097FC6@vision.arc.nasa.gov> <37BCD6CF.B7A186BC@montana.com> <5lzozmskm3.fsf@eric.cnri.reston.va.us> <00dd01beeb11$e41a68e0$f29b12c2@secret.pythonware.com> <5lso5ese9x.fsf@eric.cnri.reston.va.us>
Message-ID: <002501beeb21$5e786830$f29b12c2@secret.pythonware.com>

Guido van Rossum <guido at cnri.reston.va.us> wrote:
> > > root.protocol("WM_DELETE_WINDOW", root.quit)
> > 
> > how about making the destroy variant the
> > default?
> 
> Good idea.  I've thought of this before, and somehow never got to it.

footnote: someone just pointed out that my Tkinter
introduction says:

...

    as of version 1.63, Tkinter will not destroy the corresponding
    widget instance hierarchy, so it is a good idea to always register
    a handler yourself:

        top = Toplevel(...)

        # make sure widget instances are deleted
        top.protocol("WM_DELETE_WINDOW", top.destroy)

    Future versions of Tkinter will most likely do this by default.

    FIXME: has this been changed in Python 1.5.2?

...

</F>




From mbf2y at my-deja.com  Mon Aug 16 10:44:56 1999
From: mbf2y at my-deja.com (mbf2y at my-deja.com)
Date: Mon, 16 Aug 1999 14:44:56 GMT
Subject: Difficulty configuring web server for Python
Message-ID: <7p9855$5tg$1@nnrp1.deja.com>

My situation: I just obtained a "virtual web server" account at a web
hosting place, where I have full access to UNIX on the box and can
modify my apache config files, etc.  They advertise that they support
Perl, C, and C++ for CGI scripting.  Needless to say, first thing I did
when I got the account was to build a copy of Python.  Python appears to
work fine.

I built a simple CGI script and kept getting error 500, so I simplified
to the below script:

#!/usr/home/campusch/usr/local/bin/python


print "Content-type: text/html"
print
print "<HTML><BODY><H1>Hello World!</H1></BODY></HTML>"
print


1) Python is installed in /usr/home/..../python
2) running "python simp.py" from the command prompt gives me the
expected result.
3) running "./simp.py" from that directory runs the program just fine.
4) python itself has been chmod ugo+x so the web server process should
be able to run it.
5) This script has been chmod ugo+x so the web server process should be
able to run it.
6) The script is in the correct directory; I installed a "supported"
perl script in this directory and got it to work no problem.

This is what gets written to the server log when I try to run my script:

[Sun Aug 15 21:59:31 1999] [error] (2)No such file or directory: exec of
/usr/local/etc/httpd/cgi-bin/simp.py failed
[Sun Aug 15 21:59:31 1999] [error] Premature end of script headers:
/usr/local/etc/httpd/cgi-bin/simp.py

While the "no such file or directory" threw me, this is not a file not
found error... when I tried to run a script that really wasn't there, I
got:

[Sun Aug 15 22:10:18 1999] [error] script not found or unable to stat:
/usr/local/etc/httpd/cgi-bin/foo.py

I noticed in my srm.conf file that they had the line
AddType application/x-httpd-cgi .cgi

in there, so I renamed the .py file to .cgi.  No help at all.

As for the OS I'm running, here's what python prints on the first line
when I run it:

Python 1.5.2 (#3, Aug 15 1999, 19:04:09)  [GCC 2.7.2.1] on bsdos3

My first recourse (after spending 3 hours trying to chase this down) was
to contact tech support; they agreed with me that Python is a robust
scripting language, but in order to make sure that they have time to
handle all their clients' requests for supported software, they have had
to tighten their policy and they aren't allowed to even help me a teensy
bit when I'm using unsupported software.

Frustrating as heck.  Anyway, if anyone here has any suggestions as what
to try next, I would appreciate it.  I can't recompile apache but I can
change the config files.  I can however run a script to find out the
version of Apache I'm using, and here it is:

Server version: Apache/1.3.1.1 SSL/1.15 PHP/3.0.3
Server built:   Jun  3 1999 11:15:00
Standalone Configuration

I don't want to have to learn perl and I really don't want to do my
scripting in C++.  So if anyone here has any suggestions, I'd greatly
appreciate it.

Thanks,
-Fred


--------
Michael "Fred" Fredericks
PhD Student, Computer Science
University of Maryland College Park
fred.at.cs.dot.umd.dot.edu  ---- note I never check my dejanews mail.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From skip at mojam.com  Sun Aug 29 23:09:50 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sun, 29 Aug 1999 22:09:50 -0500 (CDT)
Subject: Parrot-0.0.1
In-Reply-To: <MPG.1233dfb6145766d998968c@news.mop.no>
References: <935279776snz@vision25.demon.co.uk>
	<MPG.1233dfb6145766d998968c@news.mop.no>
Message-ID: <14281.62669.466599.913229@dolphin.mojam.com>

    Alexander> The problem as I see it is that the descriptive language is
    Alexander> not the primary tool: once you've generated your GUI source
    Alexander> code (eg., in Python), you're separated from the original
    Alexander> script. There's no way to re-generate the GUI framework and
    Alexander> preserving your logic -- because there is no separation here
    Alexander> between presentation and code. You assume they're be
    Alexander> intermingled.

It's doable.  OMTool (Jim Rumbaugh's work at GE CRD before leaving to join
Rational) was perfectly happy to regenerate C/C++ code skeletons the user
had already modified and get the guts put back where they belonged.

One solution here would be to have two files, one that's a GUI description
and one that's an implementation file for one of the N backend languages.
The GUI description would indicate the name of the callback function and
generate code to call it.  The definition of the callback would be in an
entirely different file.  The user should never need to edit the generated
code.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From tchrist at mox.perl.com  Thu Aug 19 11:21:55 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 08:21:55 -0700
Subject: Python to Perl Conversions
Message-ID: <37bc1303@cs.colorado.edu>

Per your collective requests, here is one of the documents
with my collective observations from learning Python.

Some have mentioned that I've left things out that are in
more recent releases.  I'm sorry, but I only have Mark Lutz's
book to go on.  There are no manpages for Python.   Like most
people, I'll wait until the next edition of the book is release
rather than digging around elsewhere if the manpage are missing.

This article was posted to the Perl newsgroup, but I have duplicated
the posting here without crossposting to avoid flame wars.

--tom

                        +-----------------------+
                        | Operations on numbers |
                        +-----------------------+

Python Construct                Perl Equivalent
===========================================================================
x + y                           + for numbers, . for strings, and
                                interopolation for arrays (@a, @b)
                                and hashes (%a, %b)

x - y                           same

x * y                           * for numbers, x for strings and lists.

x / y                           Same for floats, but Perl doesn't do
                                integer division here usually.  Python
                                things 3/4 is 0 (and false), but 3.0/4.0 is 0.75

x % y                           same; Perl and Python use the same
                                definition of modulus, even though this
                                differs from that of C with respectd to
                                negative operands.

abs(x)                          same

int(x)                          same, except perl only requires it for
                                truncation toward zero, and python
                                needs it all over the place, since
                                strings aren't numbers, nor are they
                                autoconverted.

long(x)                         These are bignums, not C longs, so:
                                    use Math::BigInt;
                                    $n = Math::BigInt->new(expr);

float(x)                        not applicable

divmod(x,y)                     no direct equiv; same as
                                    ( int($x / $y), $x % y )

pow(x,y)                        x ** y

x | y, x ^ y, x & y             same, although python's only work on ints
                                and its bignum longs and blow up
                                on floats, and perl's don't work on
                                Math::BigInts, but do work on strings in
                                a very different way, and truncate floats.

x << n, x >> n                  same, although Perl works differently if
                                you go over 31 bits:  Python keeps giving 
                                you 0, whereas Perl wraps.  1<<34 is 0 in 
                                python but 4 in perl.

~x                              same-ish, but see |&^ above for details.
                                In particular, you can't negate bitstrings
                                in python.

===========================================================================

                            +----------------------------+
                            | Operations on dictionaries |
                            +----------------------------+

Python Construct                Perl Equivalent
===========================================================================

len(d)                          keys %d, in a scalar context; e.g.:
                                    $count = keys %ENV;

d[k]                            "same": $h{$k} as a real %hash, and
                                $hr->{$k} as a reference

d[k] = x                        "same": 
                                    $h{$k} = x;         # real %hash
                                    $href->{$k} = x;    # as reference
                                Note that Python doesn't support hash
                                slices, which in Perl are simple:
                                    @hash{k1,k2,k3} = (v1, v2, v3)

del d[k]                        delete $h{$k}, but Python can't do more
                                than one at a time on a hash, as in Perl's
                                    delete @hash{k1,k2,k3}

d.items()                       Just use %hash in list context to get all
                                the key/value pairs as a flat list.

d.keys()                        keys %hash

d.values()                      values %hash

d.hash_key(k)                   exists $hash{k}

===========================================================================

            +--------------------------------------------------+
            | Operations on sequences (strings, tuples, lists) |
            +--------------------------------------------------+

x in s                          No built-ins, must use grep or loop.
                                    Lists/arrays: grep { $x == $_ } @s
                                    Strings: index($s, $x) != 0

x not in s                      See previous entry.
                                    Lists/arrays: !grep { $x == $_ } @s
                                    Strings: index($s, $x) < 0

for n in x                      for $n (@array)                 # arrays
                                for $c (split //, $string)      # strings

s + s                           $s . $s

s * n, n * s                    $s x $n for strings and lists/arrays, but 
                                must be in that order in Perl.  Python
                                allows the number in either place.

s[i]                            $array[i] for arrays, but
                                substr($s, $i, 1) for strings

s[i:j]                          @array[ $i .. ($j - 1) ]  for arrays, but
                                substr($s, $i, $j - 1) for strings.  Yes, 
                                slices in python don't include the
                                termination point.

len(s)                          @array in a scalar context for arrays, e.g.
                                        $count = @array;
                                but length($s) for strings.

min(s), max(s)                  no equivalent in core perl, so you'd need a
                                loop,  although the CPAN module builtin.pm
                                provides these in fast C code.

===========================================================================

                    +---------------------+
                    | Operations on lists |
                    +---------------------+

s[i] = x                        $a[i] = expr, but Perl can do this even if
                                the array isn't long enough yet, and Python
                                can't.  You have to grow it with the append
                                method below.  

s[i:j] = x                      @a[ i .. (j - 1) = list_expression
                                but see above.  Also, Perl's slices needn't
                                be contiguous, and Python's must.  Yes, slices
                                don't include endpoint in Python.

del s[i]                        splice(@a, i, 1)

del s[i:j]                      splice(@a, i, j - 1) for values of i and j.
                                Yes, slices don't include endpoint in Python.

s.append(x)                     push(@array, listexpr)

s.count(x)                      grep { $_ == expr } @array (for numbers)
                                grep { $_ eq expr } @array (for strings)
                                grep {  /expr/    } @array (for patterns)

s.index(x)                      no direct equivalent, need loop:
                                    $found = undef;
                                    for $i ( 0 .. $#array ) {
                                        if ($array[$i] == expr) { # numeric
                                            $found = $i;
                                            last;
                                        }
                                    } 

s.insert(i,x)                   splice(@array, $i, 0) = x;  

s.remove(x)                     this deletes the first x in s, and has no
                                direct equivalent.  the simple
                                    @a = grep { !/x/ } @a
                                deletes all of them, but as a pattern.  For
                                the first, you'd need a loop or extra
                                variable, because grep doesn't shortcircuit.  
                                    $found = 0;
                                    @a = grep { !/x/ || !$found++) } @a;
                                or as a string
                                    $found = 0;
                                    @a = grep { $_ ne x || !$found++ } @a;

s.reverse                       @a = reverse @a; note that python's reverse
                                is in-place *only*, and doesn't work on
                                character strings.

s.sort(cmp?)                    Noting that Python's sort is in-place only, use
                                    @a = sort @a; 
                                    @a = sort some_cmp @a;


===========================================================================

                            +--------------------+
                            | built-in functions |
                            +--------------------+

apply(fn,args)                  &fn(@args) will skip prototype checks that
                                fn(@args) would normally enforce.  In
                                general, perl doesn't need this at all,
                                because indirect function calls don't do 
                                prototype checks.
                                    $name = "fn";   # or \&fn
                                    $name->(@arglist)

callable(x)                     n/a or else UNIVERSAL::isa($x, "CODE") 

chr(i)                          same

cmp(x,y)                        x cmp y

coerce(x,y)                     n/a!

compile(string,label,kind)      $coderef = eval "sub { $string }"

delattr(object,name)            n/a?  "delete named attribute from object".
                                I suppose this might be
                                        delete $obj->{name}
                                but that's pretty uncool to do it without 
                                asking the object.

dir(object)                     n/a for most.  returns a list of attribute
                                names in object or the caller's local scope.
                                for modules, you can do 
                                        keys %module::
                                and for hash-based objects, 
                                        keys %$obj
                                But you can't get the local (lexical
                                my variables) names in Perl without
                                resorting to C code.

eval(code)                      Same as Perl, but Perl's handles
                                statements, and python's only handles
                                expressions.  Also, python hurls an
                                exception if it doesn't like you,
                                whereas Perl sets $@.  People use this
                                to convert string to numbers.  This is
                                a terrible security hazard.

exec(code)                      eval(code);  plus see previous entry.

execfile(file)                  do 'file';  plus see anteprevious entry.

filter(func,list)               grep { func($_) } list

getattr(obj,name)               n/a?  maybe just $obj->{$name}.  I don't
                                understand why this exits.  Oh heck, yes I
                                do.  Because you can say
                                    name = "widget_count"
                                    obj.name
                                Because there's no interpolation and
                                otherwise it will look up name.  Oh my.
                                This is like the apply() mess.

globals()                       keys %{ __PACKAGE__ . "::" }

hasattr(obj,name)               exists $obj->{$name}

hash(x)                         huh?  "returns the hash value of the object,
                                if any"

hex(n)                          sprintf("%#x", n)
                                Perl's hex() function is the opposite.
                                It converts 22 into 37.  Python's hex
                                function converts 37 into 0x22.  Oh my!

id(obj)                         Just use $obj in numeric context. "returns
                                the unique id for the object, its addr in memory"

input(prompt?)                  n/a.  This reads and evals the input
                                stream!  Needed because otherwise you can't
                                read a number.  Or you could read the
                                string and convert it.  Ick.  Security
                                hazard waiting to happen.

locals()                        n/a.  We can't get our block's lexicals
                                without resorting to C.  And even if
                                we hand them, we can't symbolically
                                dereference they're names, since the
                                package (module's global) symbol table
                                is completely separate from that of the
                                lexicals' (local) one, and sym deref
                                only does the former.

map(fn, list)                   map(fn($_), list) # or written
                                map { fn($_) } list
                                Except Perl's map won't parallelize, as in
                                    map(fn, [1,2,3], [4,5,6])
                                will call fn() thrice, as in
                                        fn(1,4)
                                        fn(2,5)
                                        fn(3,6)
                                Note that Perl's map doesn't have to be
                                function; it can just be a code block.
                                    @trips = map { 3 * $_ } @singles;

oct(n)                          sprintf("%#o", $n)
                                Like hex, Python has this completely
                                backwards.  Python's oct(44) returns 054,
                                but Perl's oct(44) return 36.  UG!

open(path,mode)                 Perl's open function works very differently.
                                    $success = open(FH, mode . path)

ord(c)                          same

range(start?, end, step?)       Either use a range operator, but be careful
                                of the end point:
                                    @nums = $start .. ($end - 1) 
                                or a for() loop
                                    for ($i = $start; $i < $end; $i += $step)
                                or even 
                                    for $i ( $start .. ($end - 1) ) 

raw_input(prompt?)              print "prompt";
                                $line = <STDIN>;
                                chomp($line);    # but perl groks RS

reload(module)                  First delete from %INC, then require the
                                module again, as in:
                                    delete $INC{"Module.pm"}
                                    require Module;
                                Note that Python doesn't directly support
                                nested modules as in
                                    require Dir1::Dir2::Dir3::Module;
                                except via a disgustingly named hackaround
                                module.

repr(x)                         "$x".  Python also use `x` for this.

setattr(obj,name,value)         $obj->{$name} = $value;
                                Interpolation wins again.  See getattr.

str(x)                          I thought this was the same as repr().

tuple(seq)                      n/a.  This is a nasty hack to deal with the
                                fact that (1,2,3) is different from [1,2,3].
                                Hm... maybe it's like this:
                                    $aref = [1,2,3]; 
                                    @list = @$aref;

type(obj)                       ref($obj) returns its string type

vars(obj)                       n/a, maybe.  keys %$obj, but not really
                                anything for the locals.

xrange ...                      Like range, but avoids generating all at
                                once as Python's
                                    for i in range(100000):
                                would.  Perl's solution was to fix 
                                    for $i (1 .. 99999)
                                instead to do lazy evaluation.

===========================================================================

                    +--------------+
                    | file methods |
                    +--------------+

s = file.read(size?)            $bytes_read = read(FH, $s, $size)
                                Perl's read() doesn't let you leave
                                off the last argument to mean "read all".
                                Usually, you undefined your input record
                                separator and use readline for that.

file.readline()                 $s = readline(*FH)
                                    or more commonly
                                $s = <FH>

file.readlines()                @a = readline(*FH)
                                    or more commonly
                                @a = <FH>

file.write(string)              print FH "string"

file.writelines(list)           print FH @array_of_lines

file.close()                    close(FH)

file.tell()                     tell(FH)

file.seek(offset,whence)        seek(FH, offset, whence)

file.isatty()                   -t FH

file.flush()                    Either FH->autoflush(1) if you're using the
                                aliasing modules, or else the klunky but
                                fast:
                                    $old_fh = select(FH);
                                    $| = 1;
                                    select($old_fh);
                                Note that this just sets autoflushing
                                on each subsequent output on that handle.

===========================================================================

    +------------------------------------------------------------+
    | sys library.  There aren't there unless you have imported  |
    | the library module via "import sys".                       |
    +------------------------------------------------------------+

sys.argv                        @ARGV but sys.argv[1] in Python 
                                is $ARGV[0] in Perl, and sys.argv[0]
                                in Python is $0 in Perl, which is mutable.

sys.builtin_module_names        You either recursively go through %main::,
                                or look at %INC.  But these only show what
                                was loaded, not what was built in.

sys.exc_type                    n/a or $@.   Perl doesn't have even
                                loosely typed exceptions, just strings.
                                There is support for exception objects,
                                but people don't use them.

sys.exc_value                   The $@ variable, mostly.  See previous
                                entry.

sys.exc_traceback               n/a or $@.  We have no exception
                                traceback object.  Traceback gets appended
                                to value.

sys.exit(status)                exit(status) and no import is bloody
                                necessary just to leave the program.
                                But that Python does this by raising
                                a SystemExit exception.  Perl can use
                                END{} handlers to catch these and do
                                at-exit processing, and Python doesn't.
                                You have to roll your own.

sys.exitfunc                    This is one function to call on normal exit
                                (not exception exits, unlike perl).
                                It would have to manage all the handlers
                                in your roll-your-own scheme.  Very ugly.

sys.getrefcount(object)         This isn't exposed in Perl, save through the
                                standard Devel::Peek module.

sys.last_type                   n/a.  Type of last unhandled exception.

sys.last_value                  n/a.  Value of last unhandled exception.

sys.last_traceback              n/a.  Traceback of last unhandled
                                exception.

sys.modules                     %INC

sys.path                        @INC

sys.platform                    $^O  # that's a ^ and an O
                                You can use the $OSNAME alias from the
                                std English module if you like.

sys.ps1, sys.ps2                n/a.  interactive prompts.

sys.setcheckinterval(reps)      n/a.  This is some event hook processing
                                thing.

sys.settrace(fn)                n/a. set system traceback function.  I
                                guess you could play with the Carp module.

sys.setprofile                  n/a;  the profiler is separate in perl

sys.stdin                       STDIN

sys.stdout                      STDOUT

sys.stderr                      STDERR

sys.tracebacklimit              n/a

===========================================================================

    +---------------------------------------------------------------+
    | string library.  There aren't there unless you have imported  |
    | the library module via "import string".                       |
    +---------------------------------------------------------------+

string.atof(s)                  n/a - Perl doesn't need this.  Just use
                                the string as a float, and it is one.

string.atoi(s)                  n/a - Perl doesn't need this.  Just use
                                the string as a int, and it is one.

string.atol(s)                  Perl doesn't have built-in bignums.
                                    use Math::BigInt;
                                    $n = Math::BigInt->new($s)

string.expandtabs(s,tabsize)    From a module:
                                use Text::Tabs;
                                $tabstop = 4; 
                                @without_tabs = expand(@with_tabs); 

string.find(s,sub,start?)       index(s,sub,start?)     # no import

string.rfind(s,sub,start?)      rindex(s,sub,start?)    # no import

string.index(s,sub,st)          if (index(s,sub,st) < 0) {die "ValueError"}

string.rindex(s,sub,st)         if (rindex(s,sub,st) < 0) {die "ValueError"}

count(s, sub, start?)           You use a loop or a pattern match:
                                    $count = 0;
                                    $start = 0;
                                    while (($pos=index($s, $sub, $start) >= 0){
                                        $count++;
                                        $start += $pos;
                                    }
                                Or with regex:
                                    @matches = $s =~ /sub/g;
                                    $count = @matches;  # get scalar count
                                Or shorter:
                                    $count = () = $s =~ /sub/g;
                                If you don't want regex magic chars, use
                                    $count = () = $s =~ /\Qsub/g;

string.split(s)                 @a = split(' ', $s)     # no import

string.splitfields(s,sep)       @a = split(/sep/, $s)   # no import
                                If you don't want regex magic chars:
                                @a = split(/\Qsep/, $s) 

string.join(x)                  $s = join(" ", @x);     # no import

string.joinfields(x, sep)       $s = join($sep, @x);    # no import

string.strip(s)                 Use two substs: 
                                    $string =~ s/^\s+//;
                                    $string =~ s/\s+$//;
                                Or combined for legibility and
                                extensibility:
                                    for ($string) {
                                        s/^\s+//;
                                        s/\s+$//;
                                    }

string.swapcase(s)              $s =~ tr[a-zA-Z][A-Za-z]
                                Except this isn't locale-aware.

string.upper(s)                 uc($s)      # no import

string.lower(s)                 lc($s)      # no import

string.ljust(s,width)           sprintf("%*s", -$width, $s)
                                (no import), or use printf(), or use
                                format and write statements.

string.rjust(s,width)           sprintf("%*s", $width, $s)
                                (no import), or use printf(), or use
                                format and write statements.

string.center(s,width)          Easiest with format and write.  Could
                                hack up a (s)printf otherwise.

string.zfill(s,width)           sprintf("%0${width}d", $s)

===========================================================================

        +-------------------------------------------------------------+
        | POSIX library.  These aren't there unless you have imported |
        | the library module via "import posix".                      |
        +-------------------------------------------------------------+

posix.environ                           The %ENV hash; no import needed in Perl.  
                                        However, assignment doesn't appear
                                        to propagate to unborn children as
                                        it does in Perl.

posix.error                             This might be Perl's $! errno
                                        variable, but I'm dubious.
                                        Failed syscalls in python always
                                        raise an exception.

posix.chdir(path)                       chdir(path)     # no import in Perl
                                        Return value is success in Perl --
                                        no exception raised on failure.

posix.chmod(path, mode)                 chmod(path, mode)  # no import in Perl
                                        Return value is success in Perl --
                                        no exception raised on failure.

posix.chown(path,uid,gid)               chown(path,uid,gid) # no import in Perl
                                        Return value is success in Perl --
                                        no exception raised on failure.

posix.close(fd)                         POSIX::close(fd);  # real int, not stdio 
                                        Return value is success in Perl --
                                        no exception raised on failure.

posix.dup(fd)                           POSIX::dup(fd), or just use open as
                                        shown below

posix.dup2(fd, fd2)                     POSIX::dup2(fd, fd2), or just use the
                                        basic open with its funky dup syntax:
                                            open(HANDLE2, "<&$fd")

posix.execv(path,args)                  exec path, args;  # no import in Perl

posix.execve(path,args, env)            Just set your %ENV and then exec().

_exit(n)                                POSIX::_exit(n)

fdopen ....                             Use POSIX::fdopen or funky open:
                                            open(HANDLE2, "<&=$fd")

posix.fork()                            fork()  # no import
                                        Return value is success in Perl --
                                        no exception raised on failure.

posix.fstat()                           stat(HANDLE)    # no import

posix.getcwd()                          use Cwd;  $here = getcwd();
                                        but most folks use `pwd` in perl

posix.getegid                           $)      # not a typo

posix.getpid()                          $$      # not a typo

posix.kill(pid,sig)                     kill(sig,pid)   # no import
                                        In perl, a string is ok
                                            kill("TERM", $$);  # suicide
                                        Or multipids
                                            kill("HUP", $him, $her, @them)
                                        Return value is success in Perl --
                                        no exception raised on failure.

posix.link(src,dst)                     link($src,$dst)  # no import
                                        Return value is success in Perl --
                                        no exception raised on failure.
                                        I'll stop saying this now.

posix.listdir(path)                     opendir(DH, path);
                                        readdir(DH);            # no import

posix.lseek(fd,pos,how)                 seek(HANDLE, pos, how)  # stdio
                                        sysseek(HANDLE, pos, how)  # no stdio
                                        If you only have an fd, use
                                        POSIX::seek() or fdopen it

posix.lstat(path)                       lstat(path)             # no import

mkdir(path,mode)                        mkdir(path,mode)        # no import

open(path,flags,mode)                   sysopen(HANDLE,path,flags,mode)
                                        no import of posix needed except
                                        for flags.  use Fcntl is more
                                        widely supported for this.

(f1,f2) = posix.pipe()                  pipe(FH1, FH2)          # no import

fd = posix.popen(cmd,mod,bufsiz)        No import.  Just use regular open.
                                            open(FH, "| cmd")   # writing
                                            open(FH, "cmd | ")  # reading

posix.read(fd,n)                        POSIX::read or sysread() or usually
                                        just regular read()

posix.readlink(path)                    readline(path)          # no import

posix.rename(src,dst)                   rename(src,dst)         # no import

posix.rmdir(path)                       rmdir(path)             # no import

posix.setgid(id)                        $) = id         # not a typo

posix.stat(path)                        stat(path)      # no import
                                        But if the list return is icky, you
                                        can use the File::stat module to
                                        get by-name object values.

posix.symlink(src,dst)                  symlink(src,dst)    # no import

posix.system(cmd)                       system(cmdstr)      # no import
                                        Perl also a shell-safe version:
                                        system(arglist)

posix.times()                           times()             # no import

posix.umask(mask)                       umask(mask)         # no import

posix.uname()                           POSIX::uname()

posix.unlink(path)                      unlink(path)        # no import
                                        perl also allows a list
                                        unlink(f1,f2,f3)

posix.utime ...                         utime...            # no import

posix.wait()                            wait()              # no import

posix.waitpid(...)                      waitpid(...)        # no import

posix.write(fd,str)                     use syswrite() normally, 
                                        or maybe POSIX::write


===========================================================================
-- 
If you consistently take an antagonistic approach, however, people are
going to start thinking you're from New York.   :-)
        --Larry Wall to Dan Bernstein in <10187 at jpl-devvax.JPL.NASA.GOV>



From kernr at mail.ncifcrf.gov  Thu Aug 19 23:57:31 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Fri, 20 Aug 1999 03:57:31 GMT
Subject: No man pages? (was Re: More random python observations from a perl    programmer)
References: <199908191634.KAA01333@jhereg.perl.com> <37BC75F0.B482DA6D@starmedia.net> <37bc4352@cs.colorado.edu> <19990819123502.A28466@digex.net> <37BCA7F4.C8C30F18@fnmail.com>
Message-ID: <37bcd1c6.44274466@news.erols.com>

On Fri, 20 Aug 1999 02:57:24 +0200, Per Kistler <kistler at fnmail.com>
wrote:

>John Landahl wrote:
>
>> What's all this about no Python manpages?
>> 
>>   $ man python
>>   Reformatting page.  Wait... done
>> 
>>   Misc. Reference Manual Pages               PYTHON(10 April, 1998)
>
>It's there, but I think "perldoc perl" still has more
>to say. Why not offer all possible resources?

Lack of man-hours.  Fred's the only person (AFAIK) who is officially
working on the docs.  Doing the rest takes volunteers (hint, hint).
:-)

>Even tcl has more...
>
>-Per.

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From webmaster at chatbase.com  Sun Aug 15 21:52:46 1999
From: webmaster at chatbase.com (Tim Greer)
Date: Sun, 15 Aug 1999 18:52:46 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4ce71@cs.colorado.edu>
Message-ID: <37B76EEE.359F593A@chatbase.com>

"John W. Stevens" wrote:
> 
> > In comp.lang.perl.misc,
> >     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> > :Polymorphism requires both OO training, and discipline to use, but if
> > :used correctly, it is very powerful.
> >
> > No disagreements.
> >
> > But I can't help but wonder: Is this how you keep out 99% of the
> > accidental programmers, the ones who use Perl?
> 
> Who says I keep anybody out of anything?  I teach OO.  Not C++, or
> Smalltalk. . . OO.
> 
> > If you require OO training
> > and discipline, then you set the bar at the gate untenably high.
> 
> An opinion.  Duly noted, of course.  But still an opinion.

Just as yours is, not a very good argument, is it?

> > Perl remains proudly pedestrian in its roots.
> 
> What does that mean?

You're embarrassing yourself, you just don't realize it.
 
> > It doesn't require a Computer
> > Science degree to use.
> 
> Neither does Python.  But, why in heaven's name would you talk first
> about competence, then state it as a benefit that Perl can be used
> by the untrained?

Because it's inherently easier to get started on and have a working
program then most other languages. In a short amount of time, you can
get far, yet at the same time, you can only get "so far", until you need
to put any real effort into it, so you can learn the more complex
features of the language to your advantage, which can only be obtained
with a certain amount of comprehension and competence. How did that slip
by you? How is that different from any other language? Any idiot can
learn at least _some_ of the language and create a program to do a few
simple to intermediate tasks, but they have to posses a certain amount
of intellect to further their knowledge. 
 
> Why, in heaven's name, would you prefer Joe Blow (who has worked for
> five years as a butcher) to perform your brain surgery, to a Medical
> Doctor with a degree in Neuro Surgery?

I don't think anyone would, you're blowing that out of proportion, John.
Think about it. Why would you deny any help from "Joe Blow" to assist
you in changing your tire? That doesn't mean you'll allow him to rebuild
your engine. Refer to the above statement.

> > This, too, is a feature.  Formal training is
> > optional.
> 
>    "Formal traing is optional"
> 
> Now, what defines the difference between "formal" and "informal"
> training?

Refer to my above statements. I'm sure you'll figure it out. :-)

> And, how do you figure that OO is more difficult, more
> formal than what-ever-it-is that you are talking about?

Who said that? However, would you suggest a person start right in trying
to learn C++ and tell them how it's not important (or that it's
better/wiser) to learn C first? Taking the proper steps, learning one
thing before you learn something more indepth/complex perhaps? That's
the first thing that comes to mind for an answer when I read your
paragraph. Then again, admittingly, I'm failing to understand what you
mean to say in this post, so maybe you can clarify?
-- 
Regards,
Tim Greer   : webmaster at chatbase.com  | software at linkworm.com
The ChatBase: http://www.chatbase.com | 250,000+ hits daily Worldwide!
TRG Software: http://www.linkworm.com | CGI scripts in Perl/C, & more.
Unix/NT/Novell Administration, Security, Web Design, ASP, SQL, & more.
Freelance Programming & Consulting, Musician, Martial Arts, +Sciences.



From BC at prism.co.nz  Thu Aug 26 06:22:48 1999
From: BC at prism.co.nz (Bing Chen)
Date: Thu, 26 Aug 1999 18:22:48 +0800
Subject: Newbie: About PyRun_File()
Message-ID: <7q2mn1$sik$1@newsource.ihug.co.nz>

How can I use PyRun_File() or PyRun_AnyFile() in my application to run a
Python file? or where can I got an example?

Many thanks
BC






From moshez at server.python.net  Tue Aug 10 09:24:26 1999
From: moshez at server.python.net (Moshe Zadka)
Date: Tue, 10 Aug 1999 09:24:26 -0400 (EDT)
Subject: __init__ question
In-Reply-To: <000901bee32f$3f82fe30$3acbd9c2@peridot.optichrome.com>
Message-ID: <Pine.LNX.4.10.9908100918190.10250-100000@starship.python.net>

[Adrian Eyre, in response to the following challenge:

a = Vector()  # x=0 y=0 z=0
a = Vector( 1)  # x=1 y=1 z=1
a = Vector( 1, 2, 3)  # x=1 y=2 z=3]

How about:
 
>>> class Vector:
...  def __init__(self, *args):
...   print args
...
>>> v=Vector()
()
>>> v=Vector(1)
(1,)
>>> v=Vector(1,2)
(1, 2)


The only problem with this solution is that you have to parse
args yourself (raising TypeErrors for wrong number of args, etc.),
and having no keyword arguments, as in

Vector(z=5) # x=0, y=0

A quicker, easier and more intuitive solution is:

class Vector:

	def __init__(self, x=0, y=0, z=0):
		self.x, self.y, self.z = x, y, z


all examples from the original post, plus my own, work with the added
benefit Python does all argument parsing for us.





From kernr at mail.ncifcrf.gov  Mon Aug  2 20:58:00 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Tue, 03 Aug 1999 00:58:00 GMT
Subject: Any Python User in the Caltech Area?
Message-ID: <37a63cd3.1064954@news.erols.com>

Hello everybody,

First, let me apologize if anyone considers this message to be spam.
(Although, in this group SPAM(TM) seems to be quite popular.)

I start at Caltech in the fall, and I would like to know if Python is
being actively used by anyone at Caltech or nearby (e.g. JPL).

Thanks in advance!

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From hei at adtranzsig.de  Wed Aug 11 18:23:50 1999
From: hei at adtranzsig.de (Dirk-Ulrich Heise)
Date: Thu, 12 Aug 1999 00:23:50 +0200
Subject: Macro preprocessor.
References: <slrn7r291c.137.kaz@ashi.FootPrints.net>
Message-ID: <934412547.227655@news.adtranzsig.de>

Kaz Kylheku schrieb in Nachricht ...
>Hi folks. I wrote a macro preprocessor that can be used with a language
that
>uses indendation for statement or declaration grouping, because it is
>intelligent about dealing with leading whitespace.
>
>http://users.footprints.net/~kaz/mpp.html

Looks exactly like one of the features i missed in Python...

<duck>

Dirk






From sca at isogmbh.de  Mon Aug  9 06:46:17 1999
From: sca at isogmbh.de (Chris...)
Date: 9 Aug 1999 10:46:17 GMT
Subject: Problems with Grail 0.6 and Python 1.5.2 (NT)
References: <933940498.641759@mhs0096x.mchh.siemens.de> <7omaa2$36e$1@jaydee.iway.fr>
Message-ID: <934195578.197097@mhs0096x.mchh.siemens.de>

On Mon, 9 Aug 1999 12:29:42 +0100, Frederic Thiriet <fthiriet at coming.fr> wrote:
>
>Have you tried
>
>> python grail.py

Didn't work, either :-(



From goertzen at norscanNOSPAM.com  Thu Aug 26 12:41:03 1999
From: goertzen at norscanNOSPAM.com (Daniel Goertzen)
Date: Thu, 26 Aug 1999 16:41:03 GMT
Subject: running python on a floppy?
Message-ID: <z6ex3.1541$1i4.128395@typhoon.mbnet.mb.ca>

Do you guys think I could squeeze a linux kernel and a python interpreter
onto a floppy disk?  I'm concerned that the interpreter might use some
gigantic shared libraries ...

This will be for a web accessable embedded system.  There's plenty of ram
for ramdisks and such.

Thanks,
Dan.





From dwalker at iximd.com  Mon Aug 30 10:14:45 1999
From: dwalker at iximd.com (Damond Walker)
Date: Mon, 30 Aug 1999 14:14:45 GMT
Subject: running python on a floppy?
References: <z6ex3.1541$1i4.128395@typhoon.mbnet.mb.ca> <37CA50EC.4FC03560@home.com>
Message-ID: <plwy3.580$g57.91886@typhoon2.gnilink.net>

Doug Hellmann wrote in message <37CA50EC.4FC03560 at home.com>...
>Why are people still trying to fit distributions like this on a
>*floppy*?  Why not use a Zip disk?  Granted, they are more expensive,
>but there is *so* much more room.
>


    Sure, the ZIP is better for such things but I'd assume that most PC
style hardware comes with a 3.5" drive.  This isn't true for the Zip drive.
    Other than that, it's an exercise in pain.  And some people dig that.
:)

            Damo





From Gary.Pennington at uk.sun.com  Tue Aug 17 11:44:53 1999
From: Gary.Pennington at uk.sun.com (Gary Pennington)
Date: Tue, 17 Aug 1999 16:44:53 +0100
Subject: Problem trying to spawn (fork) asynchronous process from cgi IGNORE 
 EARLIER POST, PROBLEM SOLVED
References: <37B97F02.C5A27998@uk.sun.com>
Message-ID: <37B98375.D44F7F8@uk.sun.com>

I found the answer, I needed to close std[in,err,out] in the cgi script as
well.

Sorry for wasting bandwidth.

Gary


Gary Pennington wrote:

> Hi,
>
> I'm trying to spawn(Win32 equivalent of fork and exec) a process from
> python running under NT using CGI. I've got almost everything correct,
> but one last thing is troubling me.
>
> I invoke the script, my background process launches asynchronously(using
> hte P_DETACH option to os.spawnve), I display my message in the browser
> with successful notification - however the browser thinks that a
> transfer is still running and thus continues displaying busy. If you
> terminate the transfer, you get a transfer interrupted message in the
> browser, other than this there is no impact on the back end server
> process.
>
> It looks ugly and is bugging me, so is there a way to stop this
> behaviour? I suspected that it was caused by inherited open streams (eg
> stdin etc...,) but I closed these down in my forked program (I don't
> need them anyway) and that didn't change the problem. Interestingly, the
> browser knows that the back-end proces is still running and stops the
> transfer activity when the back end process is shut-down.
>
> Thanks for any pointers.
>
> Gary




From tbryan at arlut.utexas.edu  Tue Aug  3 21:17:21 1999
From: tbryan at arlut.utexas.edu (Thomas A. Bryan)
Date: Tue, 03 Aug 1999 21:17:21 -0400
Subject: A question of style
References: <roy-0308991545580001@qwerky.med.nyu.edu>
Message-ID: <37A794A1.90900A3F@arlut.utexas.edu>

Roy Smith wrote:

> on each page that need to access the database.  Thus, I end up
> with stuff like:
> 
> self.page.handler.server.db.cursor()
> 
> inside my widget class methods.  This works, but it's kind of 
> messy.  I'm thinking of giving my base widget class a cursor 
> method, which looks like:
> 
> def cursor(self):
>    return self.page.handler.server.db.cursor()

Either that, or you could just make an alias like
dbconn = self.page.handler.server.db
and use it like this
dbconn.cursor()

I personally like the cursor, but I thought I'd mention 
the aliasing technique in case you had forgotten about it.



From gmcm at hypernet.com  Sun Aug 29 18:07:01 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Sun, 29 Aug 1999 17:07:01 -0500
Subject: Why do we call python a scripting language?
In-Reply-To: <37c94f3a.2285568@news.netset.com>
Message-ID: <1276161941-48150991@hypernet.com>

Guy Oliver wrote:

> ...Many of these people are used to thinking of scripting
> languages as /bin/sh or batch files on DOS.  And with this mindset,
> I dont blame them for questioning my wanting to use a 'scripting'
> language on a project. Education is the only solution that there is,
> but its a slow process.

I have a client like that.

I long ago decided it was easier to get forgiveness than permission.

don't-ask-don't-tell-ly y'rs

- Gordon



From kistler at fnmail.com  Fri Aug 20 07:08:01 1999
From: kistler at fnmail.com (Per Kistler)
Date: Fri, 20 Aug 1999 13:08:01 +0200
Subject: More random python observations from a perl programmer
References: <37BC6986.A66B044E@starmedia.net> <199908191634.KAA01333@jhereg.perl.com> <012d01beeae7$f3646320$f29b12c2@secret.pythonware.com>
Message-ID: <37BD3711.E8814F70@fnmail.com>

Fredrik Lundh wrote:

> comp.lang.python is not your ordinary usenet.  please read
> the comp.lang.python posting guidelines before you post more
> crap like this.  this is not comp.lang.perl.misc.

In a way it's pleasing, that Tom invests his time in Python 
usnet. It revealed to me a lot about the state of the scripting
community, which could be a motivator for a change in some parts,
and it also showed some real technical stuff.

It is a fact in real life, that programmers of some kind are
extremely occupied with their ideas. These ideas are mostly
very good. For me it was always natural to use different tools
for reaching a goal. Only if one knows about the ideas of
others, one can incorporate them.

A clash at first, is quite normal, because many experienced
pains, when they were forced to use a differnt language against
their will, or were even compelled to read someone elses
code in an alien language. 

What I personally would like now, is a usnet discussion
between Larry Wall and Guido van Rossum, and they would swear, 
not to use any flames. If that would produce enough good and 
copious pages, one could even publish it as a book, make many 
people happy, and finally make some money:-)

Regards, Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From andrew at starmedia.net  Tue Aug 17 14:01:52 1999
From: andrew at starmedia.net (Andrew Csillag)
Date: Tue, 17 Aug 1999 18:01:52 GMT
Subject: Bug in struct module?
References: <37B8759C.D0DF1B55@bigfoot.com>
Message-ID: <37B9A390.7E5E7A1C@starmedia.net>

Hoon Yoon wrote:
> 
> Hi,
> 
> >>> calcsize('hd')
> 16
> >>> calcsize('dh')
> 10
> That seems like a bug to me.

You forgot about structure alignment.  If the machine is 1 byte-aligned,
the sizes would be the same.  From the looks of things, your machine
aligns on 8 byte boundaries (Sun machine?).

> BTW: it won't convert
> >>> unpack('dh', s[-10:])
> (2.47032822921e-323, 16424)
> >>> unpack('hd', s[-10:])
> Traceback (innermost last):
>   File "<stdin>", line 1, in ?
> struct.error: unpack str size does not match format

It shouldn't.  Since the above sizes are what they are, you'd need a
string of 16 bytes for the second call to unpack.

If you want these to work, try the network byte order/alignment option
(prefix your format string with a '>').  On my sun box, both sizes
become 10 and the unpack should also work.
-- 
"Programmers are just machines that convert coffee to executable code"
   -Unknown

"Drew Csillag" <drew_csillag at geocities.com>



From xpolakis at hol.gr  Mon Aug  9 17:06:17 1999
From: xpolakis at hol.gr (Antreas P. Hatzipolakis)
Date: 9 Aug 1999 21:06:17 GMT
Subject: Lambda Functions (fwd)
Message-ID: <xpolakis-2708561956530001@athe530-a072.otenet.gr>

FWD MESSAGE -----------------------------------------------------------------

Subject: [HM] lambda functions
From: Ivan Van Laningham <ivanlan at callware.com>
Date: Mon, 09 Aug 1999 13:47:15 -0600
To: historia-matematica at chasque.apc.org

Hi All--
I'm working on a book about the Python programming language.  Python
allows the use of lambda functions, small anonymous functions that can
speed up processing because they are effectively inlined, and often
require less typing than conventional function definitions.

Why lambda?  I realize that the term has come from functional
programming languages, and Python's inventor explicitly took lambda
from Scheme, but I'm interested in the history of the term.  Why that
particular letter?  I assume it has something to do with the lambda
calculus, but my ignorance runs deep in that area too.

Thank you for your help,
Ivan


Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70


END ------------------------------------------------------------------


Sfakianos



From timcollinsdeja at my-deja.com  Sat Aug  7 21:24:23 1999
From: timcollinsdeja at my-deja.com (timcollinsdeja at my-deja.com)
Date: Sun, 08 Aug 1999 01:24:23 GMT
Subject: I need some advice on python
References: <7oepao$6hn$1@nnrp1.deja.com> <slrn7qmqso.igt.thantos@brimstone.mecha>
Message-ID: <7oim81$qrm$1@nnrp1.deja.com>

This may just be my windows centric way of looking at things, but I was
led to believe by the agent stuff that they were exe type files for the
agent servers that wouldn't run on windows without recompiling
(Something I want to avoid).

What I need is a agent server type of thing that allows me to write
code to use it. In other words, I'd really rather concentrate on the
agent than than the agent environment....

> There's no really good reason that the Python agent code floating
> around /can't/ run on Windows; most of the networking code will work
> regardless unless they're doing really deeply fancy things.  If they
> are, you can very likely turn one of the Distributed Object harnesses
> written in Python to your use as a means of migration.  One of the
> advantages of doing so is that agent docks needn't be limited to
> /just/ Windows or Unix but can span both transparently if you're
> shipping Python bytecode around the network.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From philh at vision25.demon.co.uk  Thu Aug 12 15:17:12 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Thu, 12 Aug 99 19:17:12 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk>
Message-ID: <934485432snz@vision25.demon.co.uk>

In article <934408001snz at vision25.demon.co.uk>
           philh at vision25.demon.co.uk "Phil Hunt" writes:
> In article <37B0E67D.27F8D2ED at prescod.net>
>            paul at prescod.net "Paul Prescod" writes:
> > Phil Hunt wrote:
> > > In article <37B02965.B1FE0E8D at prescod.net>
> > >            paul at prescod.net "Paul Prescod" writes:
> > > > Have you looked into XUL,
> > > 
> > > Glade is a GUI application that allows you to build a window visually.
> > 
> > Okay, but I saw a reference somewhere to Glade having a declarative file
> > format that could be converted into code to build the GUI. I was mostly
> > talking about the declarative file format.
> 
> I think I will have to have a look at Glade. My understanding is that 
> it uses its declarative file format to save/load UI definitions that
> people are working on.
> 
> If Parrot could read/write Glade's format, that would be good, as
> then people could write a UI in Glade and export it to any
> backend that Parrot supports.

Having looked at Glade, it is now an aim of the Parrot project to support
its file format.
 
-- 
Phil Hunt....philh at vision25.demon.co.uk




From fdrake at acm.org  Fri Aug 20 11:27:16 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Fri, 20 Aug 1999 15:27:16 GMT
Subject: making python man pages looks hard
In-Reply-To: <006801beeb08$aafe2280$f29b12c2@secret.pythonware.com>
References: <1103_935134343@shoal>
	<006801beeb08$aafe2280$f29b12c2@secret.pythonware.com>
Message-ID: <14269.29652.445999.163560@weyr.cnri.reston.va.us>

Dan Connolly <connolly at w3.org> wrote:
 > The source format is LaTeX. Converting from LaTeX is hard/messy:

Dan,
  It sure is!  You'll get no argument from me on that one.

 > For example, I conjecture that it is impossible to write a program that
 > will extract the third word from a TeX document. It would be an easy
 > task for 80% of the TeX documents out there -- just skip over some
 > formatting stuff and grab the third bunch of characters surrounded by

  Not impossible, but painful enough in the general case that I have
no plans to write the code to do it (in any language!).

Fredrik Lundh writes:
 > luckily, Fred Drake has already made most of the hard work
 > here -- check the Doc/tools/sgmlconv directory:

  Dang, my secret has been found!  ;-)
  Yes, there's a good bit of potentially interesting material there.

 > not sure how incomplete, though.  I don't have things setup
 > so I can try the current release of this (from the CVS archive),
 > but maybe Fred can give us a status update?

  I don't have time today, but I'll try to send a status report to the 
Python doc-sig next week.  For anyone interested but not subscribed to 
the doc-sig mailing list, see http://www.python.org/sigs/doc-sig/ for
information.

 > anyway, if this stuff works, writing an esis2man converter
 > cannot be *that* hard.

  Parsing esis isn't hard (it's in the XML package!), and generating
man pages isn't hard ("print" is a language statement in Python,
right?).  The difficulty is the semantic mapping; what do you want on
you manpages, how do you want them organized, etc.  The existing
structure may not be trivial to transform into manpages, regardless of 
syntax.  Having a DOM instance for the library reference doesn't make
manpages a one-liner.  ;-)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From mwh21 at cam.ac.uk  Mon Aug 23 03:02:23 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 23 Aug 1999 08:02:23 +0100
Subject: UnboundLocalError (was RE: Compound Assignment Operators ( +=, *=, etc...))
References: <000301beed2c$83b6ece0$352d2399@tim>
Message-ID: <m3vha7t1xc.fsf@atrus.jesus.cam.ac.uk>

"Tim Peters" <tim_one at email.msn.com> writes:

> [Per Kistler]
> > Could you explain to me the meaning of "<wink>", please?
> > Is it a form of ":-)"?
> 
> It's closer to ;-), but "<0.1 wink>" through "<0.9 wink>" (and an
> uncountable number of others)

But only if you start posting arbitrarily long posts to c.l.py (that
argument was on this ng wasn't it?).

irrelavent-ly y'rs - Michael



From stephan at pcrm.win.tue.nl  Thu Aug 19 06:28:17 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 19 Aug 1999 12:28:17 +0200
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com> <y0jiu6edoa9.fsf@vier.idi.ntnu.no> <xj0d7wl9acd.fsf@cyclone.harvard.edu> <37BAB643.9D4E861E@strs.co.uk>
Message-ID: <m3btc412v2.fsf@pcrm.win.tue.nl>

Ian Clarke <I.Clarke at strs.co.uk> writes:

D> I did some timing tests using the code I have attached below, here are
> the results (in seconds) for sorting a list of 200 items 10 times:
> 
> Dan Schmidt's solution (using sort): 14.98
> Stephan Houben's solution          : 2.82
> Skip Montanaro's solution          : 2.736
>  (as modified by Dan schmidt)
> Magnus L Hetland's solution        : 0.032
> Skip's solution but with arrays    : 3.182
> 
> As you can see I thought that using arrays may provide an efficiency
> speed-up for Skip's algorithm, but apparently not.  Oh, I haven't really
> checked this code too carefully, so feel free to redo the tests if you
> find any bugs in the code.
> 
> Magnus L Hetland's solution wins hands down.
> 
> Ian.
> 
> ---
> import random, array, time

You should also import whrandom for Skip's solution. 

Frankly, I'm pretty much baffled by these results.
Ian's, Skip's and mine solution all use simple swaps.
However, Magnus' solution keeps removing and appending items to lists.

And now it turns out that Magnus' approach is *faster*?
How can this be?
I demand immediate explanation! <0.999 wink>

By the way, my code could be tuned just a little more:

def s2(list): 
    """Stephan Houben's solution""" 
    length = len(list)
    randrange = random.randrange
    for i in range(length): 
        j = randrange(i, length)
	tmp = list[i]
	list[i] = list[j]
	list[j] = tmp

However, it's still far from being in the same ball park
as Magnus' code.

I'm heavily disappointed in myself.

Greetings,

Stephan 



From abigail at delanet.com  Sun Aug 15 06:40:40 1999
From: abigail at delanet.com (Abigail)
Date: 15 Aug 1999 05:40:40 -0500
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4cf5f@cs.colorado.edu> <7p2lqn$2pf@dfw-ixnews7.ix.netcom.com> <slrn7rb2s2.cf9.jstevens@bamboo.verinet.com>
Message-ID: <slrn7rd6b7.a5.abigail@alexandra.delanet.com>

John Stevens (jstevens at bamboo.verinet.com) wrote on MMCLXXIV September
MCMXCIII in <URL:news:slrn7rb2s2.cf9.jstevens at bamboo.verinet.com>:
%% On 14 Aug 1999 02:55:19 GMT, Eric Bohlman <ebohlman at netcom.com> wrote:
%% >John W. Stevens (jstevens at basho.fc.hp.com) wrote:
%% >[quoting someone whose name was lost]
%% >: > It is wrong from the Larry's principle that things that behave
%% >: > differently should have different appearances.
%% >: 
%% >: Then why doesn't Perl have two different operations for
%% >: assign/reassign, vs. add to an array?  A tiny little violation of
%% >: Larry's principles, that.
%% >
%% >Nope.  "Behave differently" hear means "accomplish two different tasks."  
%% >Comparing two strings to see if they're equal lexically is not the same 
%% >task as comparing two strings to see if they're equal numerically.  
%% 
%% Right.  Do you see what I am saying?  Automatic coercion (from string to int)
%% created a problem.  The problem was dealt with by making a change that
%% impacts the language.

What's next, different types for prime numbers and non-prime numbers?

%% As well as providing a huge opportunity for defects, the complexity
%% of the language was increased.

Look. All we've seen from you here is a 4 line Python program,
riddled with bugs. And your Perl knowledge is at the same level as my
hair dressers.  All that proves is that you are in desparate need of
something that protects you from making mistakes, and that Python isn't
giving you that.

%% >Putting a particular value in a particular position in an array that 
%% >already holds a value *is* the same task as putting a particular value in 
%% >a particular position of an array that does not yet hold a value; the 
%% >only difference is in the internal details of how to accomplish it.
%% 
%% Yes, but the context of this was TC's comparison with Python's lists.
%% 
%% For an ARRAY, your argument makes sense.  For a list, it does not,
%% yet TC was taking Python lists to task for not acting like arrays.
%% 
%% On the other hand, Perl's behavior encourages defects:
%% 
%% $array[$index] = $newWord;
%% 
%% The program ran for a very long time, then the system started thrashing,
%% then it core dumped.
%% 
%% The value of $index?  Very, very large.  Perl happily tried to allocate
%% Gigabytes worth of core.

Well, then don't do that then. Djees. Consider a Python equivalent:

    while len (array) <= index + 1:   # Why isn't that array.len() ???
        array.append (0)
    array [index] = newWord;

Would that somehow magically have prevented index not to be large?

%% For example (another real bug, here):
%% 
%% $result = $Total / $count;
%% 
%% Generated a division error.  $count contained the string: "Payroll". . .


And in Python it would produce some result?



Abigail
-- 
perl -MLWP::UserAgent -MHTML::TreeBuilder -MHTML::FormatText -wle'print +(
HTML::FormatText -> new -> format (HTML::TreeBuilder -> new -> parse (
LWP::UserAgent -> new -> request (HTTP::Request -> new ("GET",
"http://work.ucsd.edu:5141/cgi-bin/http_webster?isindex=perl")) -> content))
=~ /(.*\))[-\s]+Addition/s) [0]'


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From robert_wilhelm_land at compuserve.com  Tue Aug 17 09:57:14 1999
From: robert_wilhelm_land at compuserve.com (robert_wilhelm_land)
Date: Tue, 17 Aug 1999 15:57:14 +0200
Subject: UML
Message-ID: <37B96A3A.796BB034@compuserve.com>

Are any persons in this group using the unified modelling language?
If yes, could you kindly explain when to use the UML and what the
advantages are?
I have read some articles, but I'm not able to analyse the sense of the
UML.

Would be grateful for some help


Robert





From bivey at teamdev.com  Wed Aug 18 13:38:23 1999
From: bivey at teamdev.com (William)
Date: 18 Aug 1999 17:38:23 GMT
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net> <37BAE44B.E66250BD@geo.vu.nl>
Message-ID: <01bee9a0$30018e00$583c08cf@bill.jump.net>

Patrick Bogaart <bogw at geo.vu.nl> wrote in article
<37BAE44B.E66250BD at geo.vu.nl>...
> Yes I can:
> 
> 	i = 0;
> 	label_1:;
> 	/* some use of i */
> 	i = i + 1;
> 	if (i==max_index) goto label_2;
> 	goto label_1;
> 	label_2:;
> 
> <wink>

I found this last week:

if ( something )
{
}
else
{
	A process that sets the success variable
	if (! success )
		goto Error;
}
Error:

Note that there was nothing between the goto and and the label
except the end of the else clause. -Wm




From thantos at brimstone.mecha  Sat Aug 28 15:45:01 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Sat, 28 Aug 1999 19:45:01 GMT
Subject: Returning none
References: <37C55044.8CA02530@prescod.net> <slrn7sasvb.397.kzaragoza@kzaragoza.ne.mediaone.net> <slrn7scf7d.pa.clgonsal@keeshah.penguinpowered.com> <m3wvuh8yo3.fsf@pcrm.win.tue.nl>  <slrn7sf2m1.1gv.clgonsal@keeshah.penguinpowered.com>
Message-ID: <slrn7sgesi.lbd.thantos@brimstone.mecha>

On Sat, 28 Aug 1999 07:08:05 GMT, C.Laurence Gonsalves
<clgonsal at keeshah.penguinpowered.com> wrote:

>I really *would* like to see a stronger distinction between procedures
>and functions though. I really don't think it's right for a Python

Actually, I have a solution to this that requires absolutely no code
changes, merely a perceptual change; to wit:

There are no procedures in Python.  There are only function calls (and
Zuul).

The hinge-point for understanding is that Python defines a /default/
return value for every function which is returned if there is no
explicit return; in the case of Python this value is "None".  Seen
from this perspective, you can complain about the default value for
function returns and at least be conceptually accurate but can no
longer drag a false distinction between function and procedure into
the mix.

For my part, I rather like None being the default function return
value; this lets me /treat/ a function as a procedure invisibly and
get on with things more important than jiggering with syntax, namely,
solving the problem at hand.

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From mlh at idt.ntnu.no  Tue Aug 17 18:39:43 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 18 Aug 1999 00:39:43 +0200
Subject: Unsorting(randomizing) a sequence
References: <7pb1eb$elt$1@nnrp1.deja.com> <14265.29899.470021.309355@dolphin.mojam.com>
Message-ID: <y0jk8qudob4.fsf@vier.idi.ntnu.no>

Skip Montanaro <skip at mojam.com> writes:

>     Ole> I've been writing an algorithm myself, but it's incredibly slow and
>     Ole> I would like something quicker (of course :) )

import random

result = []
for i in range(len(l)):
    element = random.choice(list)
    l.remove(element)
    result.append(element)

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From ergowolf at mindspring.com  Sat Aug 21 08:42:08 1999
From: ergowolf at mindspring.com (x)
Date: Sat, 21 Aug 1999 08:42:08 -0400
Subject: making python man pages looks hard
References: <199908202102.PAA32032@jhereg.perl.com>
Message-ID: <37BE9EA0.280D0F7C@mindspring.com>

Please contact me if you need help building the man pages.

ergowolf
ergowolf at mindspring.com

Tom Christiansen wrote:

> I've contacted the maintainer who will get back to you on it.
>
> --tom




From mal at lemburg.com  Sat Aug 21 13:44:56 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sat, 21 Aug 1999 19:44:56 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <7ple6r$kf0@ds2.acs.ucalgary.ca> <37bec222@cs.colorado.edu>
Message-ID: <37BEE598.144EBA42@lemburg.com>

Tom Christiansen wrote:
> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.python,
>     nascheme at ucalgary.ca (Neil Schemenauer) writes:
> :>    You can't use "in" on dicts.  Instead, you must use
> :In the Python philosophy, every expression should have an obvious
> :meaning.  "in" in this case does not.  Perl has a different
> :philosophy here.
> 
> To me it would be clear that it's a wasteful linear search of the
> values, just as in a list it's a wasteful linear search of the elements.
> The point wasn't a request for a feature, just an observation of a
> conterintuitive point in something that would otherwise seem to be a
> natural effect.

There are basically three "natural" approaches to "x in dict":

1. x in dict.keys()
2. x in dict.values()
3. x in dict.items()

Since dictionaries are unordered sets of relations, the third
would be the mathematically correct one... but also the most
useless one.

In Python "o in s" is only defined for sequences s with the
meaning "there is an index i so that s[i] == o". Note that there
may well be more than one such index.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   132 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From chapman at bioreason.com  Wed Aug 18 11:33:21 1999
From: chapman at bioreason.com (Mitch Chapman)
Date: Wed, 18 Aug 1999 09:33:21 -0600
Subject: waiting on a callback with PyGTK
References: <m37lmtb3sz.fsf@phaedrus.zen.xom>
Message-ID: <37BAD241.2EDD739A@bioreason.com>

Rob Hodges wrote:
> 
> I've been playing around with PyGTK and have a little program that
> pops up a dialog with a text entry in it.  When the user hits Enter in
> the text entry, a callback grabs the contents of the entry.
> 
> Now here's my problem: I want the program to sit there doing
> essentially nothing until the callback returns.

Rob,

Here's some sample code showing a solution to your problem.
The short answer is that you need a modal event loop.  (The loop
you showed would never give pygtk a chance to process any X11
events.)

BTW it might be better to route this sort of question to the
pygtk mailing list.  (But I like the fact that you're giving
exposure to pygtk here :)  See 
	http://theopenlab.uml.edu/pygtools/#pygtk-about
for subscription information.

-----------------------------------------------------------------------
#!/usr/bin/env python
"""Demonstrate the use of a modal dialog in pygtk."""

from gtk import *

class ModalDialog:
    """This is just a sample modal dialog."""
    def __init__(self, master=None):
        """Initialize a new instance.
        `master' (optional) is the GtkWindow by which this dialog is
        launched.  `master' is needed only so we can tell the window
        manager that the dialog is a transient window for some other
        window, so it should have transient window decorations."""

        self.master = master

        self.window = GtkDialog()
        if self.master:
            self.window.set_transient_for(self.master)

        # Dialogs come with containers built in: vbox for the custom
        # controls, and action_area for the dialog buttons.
        hbox = GtkHBox()
        self.window.vbox.pack_start(hbox)
        hbox.show()

        l = GtkLabel("Some Text:")
        hbox.pack_start(l, expand=FALSE, fill=FALSE)
        l.show()
            
        self.entry = GtkEntry()
        hbox.pack_start(self.entry)
        self.entry.show()

        self.okBtn = GtkButton("OK")
        self.okBtn.connect("clicked", self.OKCB)
        self.window.action_area.pack_start(self.okBtn, expand=FALSE)
        self.okBtn.show()

        # Tell Gtk+ this is a modal dialog.  Pop it up wherever
        # the mouse is at the time of popup.
        self.window.set_modal(TRUE)
        self.window.set_position(WIN_POS_MOUSE)

    def run(self):
        """Run the modal dialog.  Return whatever the user entered."""
        # This is a little weird.  Basically just run a nested
        # mainloop, and exit the mainloop by mainquit()-ing
        # out of the nested loop.  (See self.OKCB)
        self.window.show()
        mainloop()
        return self.entry.get_text()

    def OKCB(self, *args):
        """Callback invoked when the OK button is clicked."""
        # In theory, this would be a good place to validate user input
        # before committing changes.  In practice...
        self.window.hide()
        mainquit()
        

class AppWindow:
    """This is the controller for the application's main window."""
    def __init__(self):
        """Initialize a new instance -- build a UI and wire it up."""
        self.win = GtkWindow(title="Modal Dialog Example")
        # Exit when the window gets deleted/destroyed.
        self.win.connect("delete_event", mainquit)
        self.win.show()

        self.modalDialog = ModalDialog(self.win)

        # Stick a label into the window to show the results,
        # and a button which displays the modal dialog.
        self.vbox = GtkVBox()
        self.win.add(self.vbox)
        self.vbox.show()

        self.resultLabel = GtkLabel("Result:")
        self.vbox.pack_start(self.resultLabel, fill=FALSE, expand=FALSE)
        self.resultLabel.show()

        self.enterTextBtn = GtkButton("Enter Text...")
        self.enterTextBtn.connect("clicked", self.enterTextCB)
        self.vbox.pack_start(self.enterTextBtn, fill=FALSE,
expand=FALSE)
        self.enterTextBtn.show()

    def enterTextCB(self, *args):
        """Callback invoked when the 'Enter Text...' button is
clicked."""
        result = self.modalDialog.run()
        self.resultLabel.set_text("Result: %s" % result)

def main():
    """Module mainline (for standalone execution)"""
    appwin = AppWindow()
    mainloop()

if __name__ == "__main__":
    main()
-----------------------------------------------------------------------

-- 
Mitch Chapman
chapman at bioreason.com



From kperacles at geocities.com  Sun Aug  8 10:18:26 1999
From: kperacles at geocities.com (Karl Putland)
Date: Sun, 8 Aug 1999 08:18:26 -0600
Subject: Problem with modules refering to each other
References: <934075710snz@vision25.demon.co.uk>
Message-ID: <Upgr3.5987$5z4.335559@news.uswest.net>

Phil,

I agree with Klaus.  About the imports. one thing to do might be to refer to the other module only from def or class statements.

Phil Hunt <philh at vision25.demon.co.uk> wrote in message news:934075710snz at vision25.demon.co.uk...
> I have two python modules, both of which need to refer to classes 
> defined in the other one. How do I do this? 
> 
> I am currently putting at the start of each module a line of the form:
> 
>    from theOtherModule import *
> 
> However, this is not working. My modules -- greatly simplified -- look
> like this:
> 
>    #bak.py
>    from hhh import *
>    ti = AnHhhClass()
>    print ti

#bak.py
from hhh import *

def make_ti()
    ti = AnHhhClass()
    print "inside make_ti():", ti
    return ti    # If you need it outside of this function

# self test
if __name__ == '__main__':
    ti = make_ti()
    print "outside at module:", ti
 
> and:
> 
>    #hhh.py
>    from bak import *
>    class AnHhhClass: pass
> 
> When I run ``python bak.py'', I get the error message 
> ``NameError: AnHhhClass''.
> 
> -- 
> Phil Hunt....philh at vision25.demon.co.uk
> 

Karl Putland
kperacles at geocities.com




From claird at neosoft.com  Mon Aug 23 01:07:44 1999
From: claird at neosoft.com (Cameron Laird)
Date: 23 Aug 1999 00:07:44 -0500
Subject: README: Python-URL! - the weekly guide to Python resources (Aug 23)
Message-ID: <105A2F1B61569B65.A2DF0B5C14F99B47.94230FFAEC659760@lp.airnews.net>

Next week:  more working code, less territorial posturing.

    Quantitatively, the Pynews of the month was a conflict appar-
    ently about whether P-languages are in conflict.  timbot
    left no doubt about his ownership of the intellectual high
    ground with a 160-proof exposition of Perl and Python as
    comparably inferior to SNOBL.  Pathos, concision, the yacht
    race, public double dereferencing--the only thing it lacks
    is a gratuitous reference to Erlang
	http://x36.deja.com/=dnc/getdoc.xp?AN=513662476

    The great Pythonist social event of the year is the Open
    Source Software Convention, now in progress.  Early reports
    have it "well-organized"; David Ascher's tutorial was
    successful.  Look next week for more detailed summaries
	http://conferences.oreilly.com/python/

    Logistics of the Eighth PyCon are largely set.  Presenta-
    tion proposals are still being accepted
        http://www.python.org/workshops/2000-01/index.html

    A builder.com columnist reinforces the current run on
    pugnacity with his characterization of Python as "The Most
    Elegant Scripting Language You'll Probably Never Use"
        http://buzz.builder.com/cgi-bin/WebX?14 at 101.PwqganhkenJ^0 at .ee7bc67/0

    Christian Tismer shows what "dynamic module generation"
    can mean in the hands of a master
	http://x46.deja.com/=dnc/getdoc.xp?AN=512188711

    In the process of giving the one-line answer to the FAQ
    about trapping users who click on the go-away-X, Guido
    and Fredrik Lundh discover themselves trapped in one of
    their time machine's loops, as they jumble inspiration,
    intention, implementation, and documentation
	http://x40.deja.com/=dnc/getdoc.xp?AN=515152096

    Another high point of the language wars is Tom Christian-
    sen's mention of Nigel Chapman's book.  Although two
    years old and on a subject other than Python, it offers a
    lot from which we all can learn
	http://www.perl.com/language/critiques/ppc.html

    Prolific M.-A. Lemburg's mxProxy provides in its 0.2 re-
    lease not just fine-grained access control to objects,
    but also powerful weak reference methods
	http://x25.deja.com/=dnc/getdoc.xp?AN=514893032


=========================================================================
Everything you want is probably one or two clicks away in these pages:

      Python.org's Python Language Website is the center of Pythonia
          http://www.python.org

      Much of Python's real work takes place on Special-Interest Group
          mailing lists http://www.python.org/sigs/

      Python Consortium emerges as an independent nexus of activity
          http://www.python.org/consortium

      Python To-Do List anticipates some of Python's future direction
          http://www.python.org/cgi-bin/todo.py

      Python Journal is at work on its second issue
          http://www.pythonjournal.com

      Links2Go is a new semi-automated link collection; it's impressive
      what AI can generate
          http://www.links2go.com/search?search=python

    Archive probing trick of the trade:
        http://www.dejanews.com/dnquery.xp?QRY=&DBS=2&ST=PS&defaultOp=AND&LNG=ALL&format=threaded&showsort=date&maxhits=100&groups=comp.lang.python


Previous - (U)se the (R)esource, (L)uke! - messages are listed here:
  http://purl.org/thecliff/python/url.html
or
  http://www.dejanews.com/dnquery.xp?QRY=~g%20comp.lang.python%20Python-URL%21


Suggestions/corrections for next week's posting are always welcome.
  http://www.egroups.com/list/python-url-leads/

To receive a new issue of this posting in e-mail each Monday morning,
ask <claird at neosoft.com> to subscribe.


-- The Python-URL! Team--
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From philh at vision25.demon.co.uk  Fri Aug 27 17:20:22 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Fri, 27 Aug 99 21:20:22 GMT
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com>
Message-ID: <935788822snz@vision25.demon.co.uk>

In article <37c66347.518331 at news.netset.com> guy_oliver at yahoo.com  writes:
> Why do we call python a scripting language?  
> 
> For me, and for most of the people I try to tell about the benefits of
> python, the title 'scripting language' implies that its not a real
> language capable of real programs.  It implies to most of them that it
> is good for simple sysadmin and maybe short cgi scripts.
> 
> I usually simply call it an interpreted object oriented programming
> language, as this does not carry with it the implication that its not
> a real language that cant do anything.  
> 
> I usually see people grouping languages into the compiled languages,
> such as c, c++, java, ada, etc, as being the only languages viable for
> large projects, important projects.

Python and Java are the same in that they are both OO languages which 
are compiled into intermediate code which is then interpreted.

> Python, perl, tcl, and others are
> grouped into the section of scripting languages for web sites, etc.
> This is not the case, is it?

Indeed not. (Python *is* good for coding web sites -- that's one thing
I use it for -- but it's also good for lots of other stuff).

> I am not the only one that has used
> python for large projects, am I?   I hope not!  

-- 
Phil Hunt....philh at vision25.demon.co.uk




From volker.hetzer at abg1.siemens.de  Wed Aug 25 07:50:52 1999
From: volker.hetzer at abg1.siemens.de (Volker Hetzer)
Date: Wed, 25 Aug 1999 13:50:52 +0200
Subject: UML
References: <3F8BA119B48AD11197470008C7A493CB98B502@HQOBGEX1> <37c2afa8.67206110@news.compuserve.com> <slrn7s69p4.jb7.wtanksle@dolphin.openprojects.net>
Message-ID: <37C3D89C.D57E9956@abg1.siemens.de>

William Tanksley wrote:

> >I tried variuos links on the Internet, some were dead - but the most
> >led to a book advert. Two books seem to suite for my poor knowledge in
> >this area:
> 
> >1.) Object Primer(Scott W. Ambler SIGS Books/
> >     Cambridge University Press, 1995)
> >2.) UML in a Nutshell (OReilly)
> 
> >Which - or would you recommend both?
I started with "Objectys, Components and Frameworks with UML".
Was ok. Then there is "Doing hard time". This explains UML with real
objets and classes (examples come from the real-time domain).
Try to get both from a library and then pick.

> I've never seen the first, but I tried to learn from the second and HATED
> it.  The author seems to delight in using bullet lists as part of every
> paragraph, and new terms complete with definition (sometimes) are brought
> in on every page.  I was _really_ disappointed.
Nutshell books are reference books. I tried that one as well and was disappointed too.


> UML simply helps you lay out your design on paper.
Seconded. Some tools abuse UML as implementation language and spit out
C++ or something like that. Might work, might not.
Personally I'd recommend UML as a specification language.

Greetings!
Volker

-- 
Hi! I'm a signature virus! Copy me into your signature file to help me spread!



From daniel.dittmar at sap-ag.de  Tue Aug 17 06:28:58 1999
From: daniel.dittmar at sap-ag.de (Daniel Dittmar)
Date: Tue, 17 Aug 1999 10:28:58 GMT
Subject: Extension types as exceptions (Wrapup)
References: <7our22$9fb$1@mailusr.wdf.sap-ag.de> <mYDs3.1626$EG4.303931@news.shore.net> <7p8vj5$i0s$1@mailusr.wdf.sap-ag.de> <czXt3.259$mV.97228@news.shore.net>
Message-ID: <7pbdh4$kum$1@mailusr.wdf.sap-ag.de>

My code creates an actual instance of the exception class, which might
be a bit more self documenting (as the attributes have names) and is
easier to handle when there are many attributes.

This has the one disadvantage that you have to create a __str__ method
for the class (that's what the exceptionCode parameter in
createExceptionKind is for), because without it, a traceback will show
only the class name.

Actually, my current code first tries to create the instance. If this
fails, it will try to create the tuple.

There is a reference counting error in my raiseError routine:
PyInt_FromLong  (code) will set the ref count to 1,
PyObject_SetAttrString will set it to 2, so the object can never be
reclaimed.

The hopefully correct version is:

static PyObject*
 raiseCommunicationError (
    int code,
    const char * msg)
{
    PyObject * exception = PyInstance_New (CommunicationErrorType,
NULL, NULL);
    PyObject * pycode = NULL;
    PyObject * pymsg = NULL;

    pycode = PyInt_FromLong  (code);
    pymsg = PyString_FromString  (msg);
    if (exception != NULL) {
        PyObject_SetAttrString (exception, "errorCode", pycode) ;
        PyObject_SetAttrString (exception, "message", pymsg);
        Py_XDECREF (pycode);
        Py_XDECREF (pymsg);
    }
    else {
        exception = Py_BuildValue ("NN", pycode, pymsg);
    }
    PyErr_SetObject (CommunicationErrorType, exception);
    return NULL;
}


Daniel Dittmar
daniel.dittmar at sap-ag.de
SAP AG, Basis Entwicklung Berlin




From john.grayson at gsc.gte.com  Fri Aug 20 15:03:36 1999
From: john.grayson at gsc.gte.com (john.grayson at gsc.gte.com)
Date: Fri, 20 Aug 1999 19:03:36 GMT
Subject: Tkinter and popup menu
In-Reply-To: <37BD6F18.34C499AB@lri.fr>
References: <37BD6F18.34C499AB@lri.fr>
Message-ID: <7pk8q8$tp25@eGroups.com>

john.grayso- at gsc.gte.com wrote: 
original article:http://www.egroups.com/group/python-list/?start=65387
> Hi.
> 
> I want a popup menu appear when clicking with the right button on a
> canvas
> (much like a contextual menu).
> 
> how to do this ?
> 
>     stef

Here is a little fragment showing popping up a menu from an
image in a label (canvas would be similar)...

...
    xxxdata.network = Toplevel(xxxdata.root)
    xxxdata.network.title('Network Planning & Engineering')
    iimg = PhotoImage(file='%s/%s' % (xxxdata.data.bitmaps,
'network.gif'))
    ilabel = Label(xxxdata.network, image=iimg)
    ilabel.pack()
    ilabel.bind('<Button-3>', doNETWORKMenu)

...
def doNETWORKMenu(event):
    pmenu = Menu(event.widget, tearoff=1)
    # Build a regular menu, as necessary
...
    pmenu.tk_popup(event.x, event.ypos)
...





From Richard.Jones at fulcrum.com.au  Thu Aug 19 19:26:54 1999
From: Richard.Jones at fulcrum.com.au (Richard Jones)
Date: Thu, 19 Aug 1999 23:26:54 GMT
Subject: More random python observations from a perl programmer 
In-Reply-To: Message from Tom Christiansen of 1999-Aug-19 16:40:58,
            <37bc87fa@cs.colorado.edu> 
References: <37bc87fa@cs.colorado.edu>
Message-ID: <199908192326.JAA00860@envy.fulcrum.com.au>

[Tom Christiansen]
> That just triggers:
> 
>     % python
>     Python 1.5.1 (#1, Sep  3 1998, 22:51:17)  [GCC 2.7.2.3] on linux-i386
>     Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>     Traceback (innermost last):
>       File "/home/tchrist/.pythonrc", line 2, in ?
> 	readline.parse_and_bind('tab: complete')
>     NameError: readline
> 
> Time to do a real install.    Hate redHate.

   Mmmm.  Step 1 for me on a new RedHat system is almost always to replace the 
Python install with something usable like Oliver Andrich's RPMs from 
http://www.andrich.net/

   It annoys me that they ship such a minimalist version :(


       Richard





From aahz at netcom.com  Tue Aug 31 20:23:06 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 1 Sep 1999 00:23:06 GMT
Subject: Doing POST opperations from Python
References: <37cbeeee.7358361@news.freeserve.net> <14284.7351.644976.369476@dolphin.mojam.com>
Message-ID: <7qhrla$bil@dfw-ixnews8.ix.netcom.com>

In article <14284.7351.644976.369476 at dolphin.mojam.com>,
Skip Montanaro  <skip at mojam.com> wrote:
>
>    Phil> I am trying to produce POST requests within python (to drive a web
>    Phil> based email client). I can do GET requests, but can understand how
>    Phil> to do post. I have read RFC1945 (Informational), but I can find
>    Phil> the info there either.  I understand that the data is sent as part
>    Phil> of the headers, but how?
>
>Depends which module you're using.  If you use a recent version of urllib,
>you can place your parameters in a dictionary, call urllib.urlencode to
>encode it and pass the results to urllib.urlopen:
>
>    import urllib
>    paramstring = urllib.urlencode({'spam': 1, 'eggs': 2, 'bacon': 0})
>    f = urllib.urlopen("http://www.musi-cal.com/cgi-bin/query",
>		       paramstring)
>    print f.read()

I'm not positive, but I believe this appends paramstring to the URL
(through GET) instead of using POST.
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From arcege at shore.net  Wed Aug 11 17:24:40 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Wed, 11 Aug 1999 21:24:40 GMT
Subject: How do I throw an exception class from C?
References: <7os9p5$q3o$1@news.clarkson.edu>
Message-ID: <sSls3.1509$EG4.273952@news.shore.net>

Brad Clements <bkc at Murkworks.com> wrote:
: In my swig'd module  I can easily throw an exception string or tuple, but
: how do I throw an exception class?

: I want to create a new exception type that is a class, create a class object
: with some init data, then set the exception for the current thread to that
: object.

: Easy in Python, hard in C. How do I do it?

: Thanks,

You've created your exception already which is good.  Now you need to set
the exception and return NULL instead of a Python object.

  PyObject *
  MyFunction(PyObject *self, PyObject *args)
    {
      ...
      PyErr_SetString(MyException, "You can't do that");
      return NULL;
      ...
    }

Returning NULL signals that an exception has been raised, and setting
the exception as above will save the exception info (a la
sys.exc_info).  You might want to read up on
http://www.python.org/doc/current/api/exceptionHandling.html

  -Arcege




From fredrik at pythonware.com  Fri Aug 27 03:52:04 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 27 Aug 1999 07:52:04 GMT
Subject: Fuzzy string matching?
References: <199908261746.TAA25440@dionysus.fw.cuci.nl>
Message-ID: <012501bef061$10db8b30$f29b12c2@secret.pythonware.com>

Hans Nowak <hnowak at cuci.nl> wrote:
> The background is this... I want to write a deck managing program for 
> Magic the Gathering (or other CCGs or similar games). The program has 
> a collection of cards it knows (ideally, all kinds of cards in 
> existance). Now I want people to enter decks, by entering names of 
> cards. However, since some card names are quite difficult or have 
> funny characters in them, I expect that some of their input will not 
> be recognized in the card database. In such cases, I would like to 
> present the user with a list of card names which match the input 
> closely, but not entirely... strings that 'almost match'. OK, so it's 
> just a fuzzy string match. Is such a thing done in Python already? If 
> not, would it be difficult to do? (I obviously don't how to do it 
> myself otherwise I would have done so already... :-/  )

the soundex module (or a similar technique, tuned to
your dictionary) could be what you need:

import soundex
print soundex.sound_similar("fredrik", "frederick")
1
print soundex.sound_similar("fredrik", "hans")
0

</F>




From mal at lemburg.com  Mon Aug  9 13:41:39 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Mon, 09 Aug 1999 19:41:39 +0200
Subject: Licenses...
References: <37A6A2DB.9B11FB01@strs.co.uk> <37A7F3CC.583F2C7F@houbits.com>
	 <37AA804D.9A1317C0@houbits.com> <934115482snz@vision25.demon.co.uk>
	 <37ae686e.59511102@news.ricochet.net> <37AE9AC3.34007C4E@lemburg.com>
	 <37AE9B42.4DF40758@lemburg.com> <SpY1dCA7Ctr3EwsH@jessikat.demon.co.uk>
Message-ID: <37AF12D3.71240BA7@lemburg.com>

Robin Becker wrote:
> 
> In article <37AE9B42.4DF40758 at lemburg.com>, M.-A. Lemburg
> <mal at lemburg.com> writes
> ...
> >If you link against a GPLed product (even if it's done dynamically!),
> >the whole construction (your product + the GPLed one) automatically
> >fall under the GPL. This means that you can still sell and distribute
> >your product, but only if you include the full source code of
> >the GPLed product
> >*plus* your product.
> >
> Oh dear; I wrote this extension called dll for Tcl that allows tcl code
> to link to arbitrary dll's. The source code for the extension is
> available. Must I distribute source for all known GPL'd products which
> can be linked to tcl using this extension. As it happens this thing is
> mostly used by win32ers, but it has run under Linux.

If TCL were under GPL then you'd have to ship both TCL and
your extension under GPL. But you're lucky: TCL is shipped under
a more liberal license, so you can distribute your code in any
form you like with or without TCL, as binary or source.

Note that I just posted this because more and more GPLed extensions
for Python are popping up -- if you want to include these in a
product, you'll have to take some care to follow the extension's
license. Otherwise, including the extension's code would be
simply illegal. In most cases, the author will probably also
distribute his/her code under a different (payed) license. In some
cases (like the GNU readline module, for which Python includes
an extension), the authors will probably not... e.g. shipping
Python with linked in readline support (as DLL or statically)
causes that Python interpreter to fall under GPL and with it,
all extensions linked to it. The Python code run by the interpreter
is not forced under GPL (otherwise all gcc compiled software
would turn GPLed...).

> >The LGPL is a little different: it allows you to link against the
> >LGPLed product without the need to ship your source code. You still
> >have to include the source code of the LGPLed product you're linking
> >against though (including any changes that you have made to it).
> >
> >If you are using Python extensions written by different authors,
> >be sure to check their license statement. In case you have any
> >doubts get in touch with them before using their code. Most
> >extensions are under very similar licenses as Python itself,
> >but some a GPLed and other LGPLed. Also, some are not valid for
> >commercial use meaning that you'll have to negotiate a special
> >license with the author to be able to use in your code. This
> >usually involves paying a license fee.
> >
> >Hope this clarifies things a little. More infos on GPL and LGPL
> >should be available on www.gnu.org. If you are still unsure what
> >is legal and was is not, go on the safe side and contact your
> >lawyer to get things straight.

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   144 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From nospam at bitbucket.com  Mon Aug 16 03:18:29 1999
From: nospam at bitbucket.com (Phil Mayes)
Date: Mon, 16 Aug 1999 00:18:29 -0700
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> <934326918snz@vision25.demon.co.uk> <37B4500C.276AD22B@basho.fc.hp.com> <37b6dcae.342432061@news.triode.net.au> <37b6f43e.1175088@news.erols.com>
Message-ID: <xXOt3.327$Vn5.34090@newsfeed.avtel.net>

Robert Kern wrote in message <37b6f43e.1175088 at news.erols.com>...
>On Sun, 15 Aug 1999 15:36:39 GMT, skaller at maxtal.com.au (John (Max)
>Skaller) wrote:
>
>[snip]
>
>>This last point is the real killer. I have both a Linux and
>>NT box. I have no problem with building C modules
>>on Linux, but I don't _have_ a compiler on the NT box that
>>will build C modules compatible with the Python implementation
>>I have. [I have Borland C++, and gcc on the NT box,
>>but not MSC, which was used to build my NT based Python]
>
>You can use GCC 2.95 for mingw32
>(http://www.xraylith.wisc.edu/~khan/software/gnu-win32/gcc.html)
>to compile Python extensions on NT.  See my page
>http://starship.python.net/crew/kernr/mingw32/Notes.html
>
>Such extensions will work with the distributed Python interpreter for
>Win32.

There is also LCC-Win32 at http://www.cs.virginia.edu/~lcc-win32/
I haven't played with it recently but I used it to teach a C course
last year and it understood windows.h and .rc files very well.

--
Phil Mayes    pmayes AT olivebr DOT com
Olive Branch Software - home of Arranger
http://www.olivebr.com/







From philh at vision25.demon.co.uk  Sun Aug 22 16:43:39 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Sun, 22 Aug 99 20:43:39 GMT
Subject: Parrot-0.0.1
References: <935279776snz@vision25.demon.co.uk> <7pp517$dce$1@nnrp1.deja.com>
Message-ID: <935354619snz@vision25.demon.co.uk>

In article <7pp517$dce$1 at nnrp1.deja.com>
           schorsch at schorsch.com "Georg Mischler" writes:
> philh at vision25.demon.co.uk wrote:
> > To use Parrot, first you create a *.par file describing your
> > application's GUI. Your file might look something like this:
> >
> > -----------Cut here-------------------------------
> > window @MyWindow "My First Window" {
> >    menuBar {
> >       menu "File" {
> >          menuItem @New "New"
> >          menuItem @Exit "Exit"
> >       }
> >       menu "Help" {
> >          menuItem @About "About..."
> 
> Looks very interesting.
> And actually reminds me a lot of autodesks slowly deprecated
> dcl description language which I also liked a lot at the time.
> Is this a coincidence?

Yep, I've never heard of Autodesk DCL.
 
> Another thing: Do you have any concepts on supporting dynamic
> dialogs with your tool?

It's not something I've thought about, up til now.


-- 
Phil Hunt....philh at vision25.demon.co.uk




From thomas at xs4all.nl  Mon Aug 23 03:52:17 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Mon, 23 Aug 1999 09:52:17 +0200
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer)
In-Reply-To: <000201beed2c$82aa5ee0$352d2399@tim>
References: <19990822210143.A706@xs4all.nl> <000201beed2c$82aa5ee0$352d2399@tim>
Message-ID: <19990823095217.D28043@xs4all.nl>

On Mon, Aug 23, 1999 at 01:58:23AM -0400, Tim Peters wrote:

> > Out of curiosity, is the compiler smart enough to see that
> > exec statements like
> > exec "a = 1 * 10" in {}

> No.  It would first have to parse the exec'ed string at compile-time, but
> that's delayed to run-time, too late to help the local/global decisions made at
> compile-time.  In real execs, one or more of the arguments are unknown at
> compile-time anyway (nobody would write the above in a real program, but if
> you're the exception feel free to submit a tedious patch to look for stuff like
> that <wink>).

Heh, like i said, it was curiosity. I can imagine code doing the above to
provide simple syntax checking before writing the exec'd code to a .py file
for use later. However, I dont plan on writing anything that does that
without using 'exec' in other places, and cares about speed. I'll use the
extra runtime to make loving python-remarks to the perl-loving colleague who
sits across from me ;-)

Nevertheless, compiler and language internals fascinate me. As do computer
internals in general. Which reminds me... I hardly see any
'development-traffic' on this list, other than some sideways references to a
CVS tree... Is there a secret society or do you just have meetings in
guido's favorite pub ?

Noisily-and-nosey-y'rs, Thomas

-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From connolly at w3.org  Tue Aug 17 00:58:43 1999
From: connolly at w3.org (Dan Connolly)
Date: Mon, 16 Aug 1999 23:58:43 -0500
Subject: if you want POSIX, import posix, not os
Message-ID: <37B8EC03.835F8A51@w3.org>

I have thought this was broken for years, but I just read
it again tonight, and I'm not going to pass over it in silence this
time:

"Do not import this module directly. Instead, import the module os,
which provides a portable version of this interface."
	-- http://www.python.org/doc/1.5.2p1/lib/module-posix.html

I think this is terrible advice.

The os module is *not* portable:

"getpgrp () 
    Return the current process group id. Availability: Unix. "
	-- http://www.python.org/doc/1.5.2p1/lib/os-procinfo.html

on my NT box:

>>> import os
>>> os.getpgrp()
Traceback (innermost last):
  File "<interactive input>", line 0, in ?
AttributeError: getpgrp

What's portable about that?

A portable interface is one where I can expect consistent behaviour
of named objects across platforms, no?

All I know about a python program that imports os is that
it might work on some platforms and not on others. That's
the exact *opposite* of portable, no?

The IEEE POSIX spec is widely implemented and understood.
When a program expects behaviour as specified in the POSIX
specs, it should say
	import posix
no? If an implementation on WinNT provides posix semantics,
then it may supply a module by that name that works as
expected. Otherwise, I want an early ImportError, not
later AttributeErrors, or worse: silent failures to
behave as specified.

It would have made sense for python to define a really portable os
module, i.e. the least common denominator of the platforms it
runs on. But that's not what we've got.

I hope the os module will be deprecated in future releases.

-- 
Dan Connolly, W3C
http://www.w3.org/People/Connolly/



From don_b at my-deja.com  Tue Aug 17 15:11:13 1999
From: don_b at my-deja.com (Don)
Date: Tue, 17 Aug 1999 19:11:13 GMT
Subject: Threads Python OpenVMS
Message-ID: <7pcc46$ej7$1@nnrp1.deja.com>

Has anyone been able to successfully build threads into Uwe Zessin's
OpenVMS port of Python-1_5_2?  I have OpenVMS/Alpha 7.2.  I am having
trouble compiling @DCC_PYTHON THREAD.  I'm getting lots of errors in
THREAD_PTHREADS.H.  I'd appreciate any input.

Thanks


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From m.faassen at vet.uu.nl  Thu Aug 12 04:08:41 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 12 Aug 1999 08:08:41 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B02965.B1FE0E8D@prescod.net> <934326002snz@vision25.demon.co.uk> <7oraiv$nuo$1@newshost.accu.uu.nl> <934408038snz@vision25.demon.co.uk>
Message-ID: <7otve9$d1u$1@newshost.accu.uu.nl>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
> In article <7oraiv$nuo$1 at newshost.accu.uu.nl>
>            m.faassen at vet.uu.nl "Martijn Faassen" writes:
[I point Phil to a page about acquisition]

> I really will have to learn Zope one of these days.

This page isn't about Zope, but about acquisition, which describes
a technique that is used in Zope, but can be used in other places
as well.

Regards,

Martijn




From pointal at lure.u-psud.fr  Mon Aug 30 04:56:19 1999
From: pointal at lure.u-psud.fr (Laurent POINTAL)
Date: Mon, 30 Aug 1999 08:56:19 GMT
Subject: building Python+readline on Suse 6.0
References: <37C9A580.A3A3D95A@gnet.tn>
Message-ID: <37ca4709.7263604@news.u-psud.fr>

On Sun, 29 Aug 1999 23:26:25 +0200, MB <ritm at gnet.tn> wrote:

>: The message::
>: -----------------
>: gcc -Xlinker -export-dynamic python.o \
>:                 ../libpython1.5.a -L/usr/lib/termcap -lreadline
>: -ltermcap
>:                  -ldl   -lm  -o python
>: /usr/i486-linux/bin/ld: warning: libc.so.6, needed by
>: /usr/lib/termcap/libtermcap.so, may conflict with libc.5.so
>: -----------------
>: 
>: how to solve it ?

? Recompile readline...

A+

Laurent.




From mlh at idt.ntnu.no  Sat Aug 28 11:28:55 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 28 Aug 1999 17:28:55 +0200
Subject: Fuzzy string matching?
References: <y0j7lmiuyo2.fsf@vier.idi.ntnu.no> <19990827231353095.AAA161.428@max41051.izone.net.au>
Message-ID: <y0jso53udoo.fsf@vier.idi.ntnu.no>

"John Machin" <sjmachin at lexicon.net> writes:

> On 26 Aug 99, at 21:31, Magnus L. Hetland wrote:
> 
> > mlh at idt.ntnu.no (Magnus L. Hetland) writes:
> > 
[...]
> 
> Magnus,
> One presumes that your first assignment for your students is to go to 
> the library and/or think a little and come back with the well-known 
> O(min(m,n)) instead of O(mn) space requirement optimisation to the 
> algorithm you presented and therefore you would prefer not to have it 
> posted to this list, just in case they read it.

Nope. I can only plead ignorance here... :)

(I'm not really the lecturer - just responsible for the exercises,
so... ;)

> Oh and for what it's 
> worth, although the time is still O(mn), it decreases --- I have yet to 
> see a language where foo[i,j] takes no more time than bar[j]. Python, 
> with foo being a dictionary and bar being a list, is definitely not 
> such a language.

I see.

I didn't know *that* much about this algorithm, really - not even that
the measure was called Levinshtein distance - until I did some
research on the net... (Nice thing that, the net :)

Found some nice stuff at

  http://www.dir.univ-rouen.fr/~charras/seqcomp/

which had several algorithms for sequence comparisons (useful for
fuzzy matching). I didn't find any description of how to reduce the
space requirements, thoug... So I had to figure that out myself ;)

A new and improved version:

def distance(a,b):
    n = len(a); m = len(b)
    if n>m: a,b,n,m = b,a,m,n
    c = []
    for i in range(0,n+1):
        c.append(i)
    for i in range(1,m+1):
        p = c; c = [i]
        for j in range(1,n+1):
            x = p[j]+1
            y = c[j-1]+1
            if a[j-1] == b[i-1]:
                z = p[j-1]
            else:
                z = p[j-1]+1
            c.append(min(x,y,z))
    return c[n]


Could perhaps be a bit more readable, though... (Maybe I'll work on
that...)

> Cheers,
> John
> 

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From jand at easics.be  Thu Aug 19 12:13:28 1999
From: jand at easics.be (Jan Decaluwe)
Date: Thu, 19 Aug 1999 18:13:28 +0200
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <37BC2D28.46D1FA8E@easics.be>

Tom Christiansen wrote:
> 
> DISSIMILARITY:
>     Because there are no variable markers in Python, there are no
>     variable interpolation in strings.  That makes something like
>         print "this $x is $y here\n"
>     become in python
>         print "this %d is %d here\n" % (x, y)

However, if you "supply" a scope, e.g. using the vars() function,
you can say:

  print "this %(x)d is %(y)d here" % vars()

which lets you combine interpolation with formatting directives.

> GOTCHA: (medium)
>     All ranges are up to but *not including* that point.  So range(3)
>     is the list [0,1,2].  This is also true in slices, which is the
>     real gotcha part.  A slide t[2:5] does not include t[5] in it.

I always found this very strange. However, in practical use it turns
out to be often quite natural, for example:
   a[:5] returns 5 elements.

Exactly the same approach is used by the containers in C++'s STL
library, so I guess there must be some good theoretical arguments also.

-- 
Jan Decaluwe	       Easics              
Design Manager         System-on-Chip design services  
+32-16-395 600	       Interleuvenlaan 86, B-3001 Leuven, Belgium
mailto:jand at easics.be  http://www.easics.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 2730 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://mail.python.org/pipermail/python-list/attachments/19990819/d97eae5e/attachment.bin>

From skip at mojam.com  Tue Aug 31 20:42:48 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 31 Aug 1999 19:42:48 -0500 (CDT)
Subject: Doing POST opperations from Python
In-Reply-To: <7qhrla$bil@dfw-ixnews8.ix.netcom.com>
References: <37cbeeee.7358361@news.freeserve.net>
	<14284.7351.644976.369476@dolphin.mojam.com>
	<7qhrla$bil@dfw-ixnews8.ix.netcom.com>
Message-ID: <14284.30272.128542.213904@dolphin.mojam.com>

    >> import urllib
    >> paramstring = urllib.urlencode({'spam': 1, 'eggs': 2, 'bacon': 0})
    >> f = urllib.urlopen("http://www.musi-cal.com/cgi-bin/query",
    >> paramstring)
    >> print f.read()

    Aahz> I'm not positive, but I believe this appends paramstring to the URL
    Aahz> (through GET) instead of using POST.

Nope.  Check the urllib docs and try the above code.  It accesses my query
cgi script.  Check the REQUEST_METHOD environment variable the script spits
out.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...




From paul at prescod.net  Sat Aug 28 06:06:36 1999
From: paul at prescod.net (Paul Prescod)
Date: Sat, 28 Aug 1999 10:06:36 GMT
Subject: Returning none
References: <000101bef12c$19566520$ad2d153f@tim>
Message-ID: <37C7B4AC.3EA833B0@prescod.net>

Tim Peters wrote:
> 
> > Personally, I think falling out of functions and expecting None to be
> > implicitly returned is a very Perl-ish thing to do...
> 
> Oddly enough not.  In the absence of an explicit "return", Perl returns the
> value of the last expression evaluated before falling off the end, so in your
> function above it returns what you intended.

It's Perl-ish in that it is trying to guess what you meant.

> Python doesn't return None for the convenience of people too lazy to explicitly
> return None themselves; it's instead a cheap way to make it probable that
> someone using a procedure as a function will get an error (None supports no
> operations except true/false interpretation) without blowing up the interpreter
> (do this in C and you'll access random stack trash).

I'm not clear what you are trying to say here. Explicitly triggering a
runtime error would not blow up the interpreter.

 Paul Prescod



From poinot at onera.fr  Tue Aug 31 03:45:23 1999
From: poinot at onera.fr (Marc POINOT)
Date: Tue, 31 Aug 1999 09:45:23 +0200
Subject: lex / yacc for Python
References: <7qfvct$o8$1@news.is.co.za>
Message-ID: <37CB8813.3509910A@onera.fr>

Jacques Oosthuizen wrote:

> Any one know if there is such a beast
>
> --
> http://www.python.org/mailman/listinfo/python-list

There is at least one parser we are using. It does both lexical
and syntaxic jobs.

http://www.chordate.com/kwParsing/index.html

It's part of Gadfly, but actually we are using it
for the parser part only.

Marcvs [alias There are other parsers, such as PyLR
but I didn't use them.]






From barmeier at mamba.com  Thu Aug 12 14:14:12 1999
From: barmeier at mamba.com (Matthias Barmeier)
Date: Thu, 12 Aug 1999 18:14:12 GMT
Subject: Copying Huge Files ? (Newbie!)
Message-ID: <37b30e52.43353218@news.usenetserver.com>

Hi,


I want to copy some files across the disks of my server using python
parse a script and then do what I want.

But while reading the manual I cannot find something like os.copyfile
or os.movefile.

Are there some functions except open/read/write that can do the copy
for me ???



Ciao
	Matthias Barmeier



From tratt at dcs.kcl.ac.uk  Tue Aug 10 06:13:44 1999
From: tratt at dcs.kcl.ac.uk (Laurence Tratt)
Date: Tue, 10 Aug 1999 11:13:44 +0100
Subject: Hidden CGI fields in HTML
References: <37A94EDA.81644EF3@NOSPAM.strs.co.uk>
Message-ID: <8b98572f49.tratt@btinternet.com>

In message <37A94EDA.81644EF3 at NOSPAM.strs.co.uk>
          Ian Clarke <I.Clarke at NOSPAM.strs.co.uk> wrote:

> I am using CGI with Python and am trying to find a way that I can generate
> the HTML for a form such that one of the INPUT controls will have a fixed
> value (set when the page is generated by a previous CGI script), that
> CANNOT be edited by the user (or prefereably not even seen) - but will be
> passed to the next CGI script.

As you have realised, this is nothing whatsoever to do with Python. Use a
type of 'hidden' in the HTML input tag.


Laurie
-- 
http://eh.org/~laurie/comp/python/



From bradh at mediaone.net  Thu Aug 19 21:18:52 1999
From: bradh at mediaone.net (Brad Howes)
Date: 19 Aug 1999 21:18:52 -0400
Subject: How to sell Python to your boss?
References: <ka8yaf7d84t.fsf@delightful.cs.hut.fi>
Message-ID: <wjywvur2qrn.fsf@bradh.ne.mediaone.net>

Petri Mikael Kuittinen <eye at niksula.hut.fi> writes:

> For several months I have tried to get my colleagues excited to learn
> Python, but so far nobody has been interested to learn new languages,
> except for some mild interest towards Java (which is IMHO an
> overly-hyped language).

Heck, I gave up and joined a company (moving ~2800 miles) that hired me
*because* of my Python skills.

YMMV... :-)

-- 
Brad Howes                    bradh at mediaone.net



From bruce_dodson at bigfoot.com  Tue Aug 10 20:01:53 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Tue, 10 Aug 1999 21:01:53 -0300
Subject: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au>
Message-ID: <q43s3.58867$jl.37270547@newscontent-01.sprint.ca>

Ni!  I don't see the point of creating a variant of python that aims to be
more efficient, but does so at the cost of taking away Python's most
powerful optimization mechanism: external C modules.  I also think that a
python that has enough knowledge of types to perform inlining... well, such
a beast is no longer Python, no matter how similar the syntax appears.




From r.skowronek at fz-juelich.de  Fri Aug 20 09:23:34 1999
From: r.skowronek at fz-juelich.de (r.skowronek at fz-juelich.de)
Date: 20 Aug 1999 13:23:34 GMT
Subject: Reading from serial port and dumping to console
References: <37BD20F8.9A984D7E@pc.jaring.my>
Message-ID: <7pjksm$4iv@zam201.zam.kfa-juelich.de>

Thomas Lim <englim at pc.jaring.my> wrote:

: I'm trying write a program to read from the serial port (barcode reader
: attached) and dump to the console, so that my application program can
: receive the data.
: OS: Red Hat Linux 6
: Python: 1.5.2

: Program listing:
: p = open('/dev/ttyS0', 'r')
: t = open('/dev/tty0', 'w')
: while 1:
:     x = p.readline()
:     print x   # yes, I could read from /dev/ttyS0
:     t.write(x)  # but this has no effect on console


I use Python to read out a multimeter and append the data to a logfile.
The serial port was initialized with Kermit before though:

#!/usr/local/bin/python
# UDL45 auslesen und Werte in File 'hv-log' schreiben (append)
import time
udl = open('/dev/cua1','r+')
log = open('hv-log','a')
udl.write('val?\r')
val = udl.readline()
t = time.time()
time.ctime(t)
#print val
log.write(time.ctime(t)),
log.write(': ',)
log.write(val)
udl.close()
log.close()


Rolf



From arcege at shore.net  Thu Aug 12 13:59:46 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 12 Aug 1999 17:59:46 GMT
Subject: Extension types as exceptions
References: <7our22$9fb$1@mailusr.wdf.sap-ag.de>
Message-ID: <mYDs3.1626$EG4.303931@news.shore.net>

Daniel Dittmar <daniel.dittmar at sap-ag.de> wrote:

: Should it be possible to use extension types (implemented in C) as
: exceptions? 

You do not want to use types as exceptions, you want to use either
strings or exceptions.

: Currently, this works partly, as you can 'raise' and 'except' them.
: It doesn't work in the following code:

: try:
:     ... code raising exception
: except EOFException, err:
:     pass

: Only strings and python classes seem to be allowed.

And C exception classes created with PyErr_NewException().  This
creates a subclass of "Exception" which can be used to raise and catch
exceptions.

You might want to read:
http://www.python.org/doc/current/api/exceptionHandling.html and
http://starship.python.net/crew/arcege/extwriting/pyext.html (under the
section "Making an exception").

  -Arcege




From skip at mojam.com  Thu Aug 26 15:52:24 1999
From: skip at mojam.com (Skip Montanaro)
Date: Thu, 26 Aug 1999 14:52:24 -0500 (CDT)
Subject: Library philosophy
In-Reply-To: <37C57A80.D5C62704@prescod.net>
References: <37C57A80.D5C62704@prescod.net>
Message-ID: <14277.39227.719967.140778@dolphin.mojam.com>

    Paul> I believe that Python's libraries are in many places too thin of a
    Paul> layer on top of the C library and are thus not very
    Paul> Pythonish. Evidence of this includes indecipherable function names
    Paul> like strtfmt and APIs that return tuples instead of objects and
    Paul> take integers as "flags" instead of strings or objects.

This is a known issue.  I'll put words in g--d-'s mouth (and let him take
them out when he returns) once again and indicate that the thin libraries
were made that way on purpose.  I can't pretend to understand all the
reasons that is the case, but it appears to be a pretty consistent approach.
It also makes the most sense to me when you are exposing a C API with an
extension module.  If you come up with some higher-level classes (e.g. a
logical filesystem that doesn't expose chmod and friends and that acts
uniformly across different OS platforms), my guess is that it would be
looked upon favorably.

    Paul> Would it it make sense to start wrapping the low-level modules
    Paul> with high-level, easier to use modules that would be "the default"
    Paul> in Python 2?

Makes good sense to me.

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...




From 71351.356 at compuserve.com  Tue Aug 17 00:46:58 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Tue, 17 Aug 1999 00:46:58 -0400
Subject: stat, chmod and RDONLY file (win32)?
References: <7pambk$3m8$1@ssauraaa-i-1.production.compuserve.com>
Message-ID: <7papne$51v$1@ssauraaa-i-1.production.compuserve.com>

Reply to myself.
I know that I can run a dos command such as ATTRIB /S -R *.* to accomplish
this.  But, that just delays having to know how the chmod works and is not
too portable.   I tried os.chmod( filename, os.O_RDWR), but it does nothing.

T. C. Mits <71351.356 at compuserve.com> wrote in message
news:7pambk$3m8$1 at ssauraaa-i-1.production.compuserve.com...
>
> Hitting a lot of stumbling blocks.   Latest is trying to delete a
directory
> tree that may have files with a read-only attrib (in NT and win98).
The
> exception occurs in shutil.rmtree(), so I guess I'll have to cobble
> something to find such files and chg the attributes.   I found os.stat(),
> chmod() and stat module, but the docs are a little shallow.







From jstevens at basho.fc.hp.com  Fri Aug 13 23:04:55 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:04:55 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132312.RAA07056@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:12 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cec7@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :> But the point is that Python refused to autoallocate for you.
> :
> :Excuse, but Python did indeed automatically allocate on:
> :
> :a.append( b )
> :
> :Will automatically allocate a space for a new element in the container
> :class object 'a'.
> 
> Sometimes it autoallocates on assignment.  Sometimes it doesn't.
> This is obviously inconsistent.

So much for freedom. . . and this from a Perl guru!

The differences are based on the differences between an object of
one class, vs. an object of another class.

By your reasoning, Perl is very, very inconsistent (since not every
data type supports '+').

> Perl, however, is perfectly consistent
> here across all assignment operations, and Python isn't.

Which represents a flaw in Perl.  Different classes should/must
treat "assignment" in different fashion.

> It's as clear
> as that.

Yes.  It is.

John S.




From doko at cs.tu-berlin.de  Wed Aug 25 09:29:37 1999
From: doko at cs.tu-berlin.de (Matthias Klose)
Date: 25 Aug 1999 15:29:37 +0200
Subject: Python parser generators
References: <934154252snz@vision25.demon.co.uk>
Message-ID: <78wvuk0z0e.fsf@cs.tu-berlin.de>

philh at vision25.demon.co.uk (Phil Hunt) writes:

> What parser generators are available for Python? I'm current using 
> PyLR, are there any others that people recommend?

TRAP A Python-based generic compiler prototyping system

	http://www.first.gmd.de/smile/trap/




From mhagger at blizzard.harvard.edu  Fri Aug  6 13:41:15 1999
From: mhagger at blizzard.harvard.edu (Michael Haggerty)
Date: 06 Aug 1999 13:41:15 -0400
Subject: Stupid Question?
References: <Bioq3.214$6E5.3247@typ12.nn.bcandid.com>
Message-ID: <xj0907oremc.fsf@cyclone.harvard.edu>

"jim moore" <indiana at ev1.net> writes:

> Is there a bulit in function that will give you N evenly spaced
> numbers between X1 and X2.  For example: 10 numbers between 0 and 1
> ([0.1, 0.2, 0.3, ...])

Here's what I use (requires Numeric).  To get `10 numbers between 0
and 1' you type `raster(0,1,10)', but note that it returns the more
pythonic [0, 0.1, 0.2, ..., 0.9].  To get your list you would type
`raster(0,1,10,1,11)'.  You can also get this as a module, through
http://monsoon.harvard.edu/~mhagger/download/.

Michael

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

import Numeric

def raster(d0, dn, n, lo = 0, hi = None):
    """Return a Numeric array of equally-spaced floating-point numbers.

        d0,dn : any numbers (real or integer)
        n : nonzero integer
        lo, hi : integers

    Return a numeric array containing values chosen by dividing the
    range from d0 to dn into n equal intervals.  By default return n
    points, including d0 but excluding dn in the usual python style.
    If lo and/or hi are specified, then select the following subset of
    (hi-lo) points from the same grid:

        [d0 + lo*delta, d0 + (lo+1)*delta, ..., d0 + (hi-1)*delta]

    where delta is (dn - d0)/n.  The idea is that you can use round
    numbers for d0, dn, and n, to get round numbers out, regardless of
    your choice of lo and hi.

    """

    delta = (float(dn) - float(d0)) / n
    if hi is None: hi = n
    return d0 + delta*Numeric.arange(lo, hi, typecode=Numeric.Float)


-- 
Michael Haggerty
mhagger at blizzard.harvard.edu



From aahz at netcom.com  Sun Aug 22 11:11:07 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 22 Aug 1999 15:11:07 GMT
Subject: Unsorting(randomizing) a sequence
References: <000401beebac$7a89b520$0c2d2399@tim>
Message-ID: <7pp3ub$gdj@dfw-ixnews7.ix.netcom.com>

In article <000401beebac$7a89b520$0c2d2399 at tim>,
Tim Peters <tim_one at email.msn.com> wrote:
>[Tim, on the difficulty of generating random permutations, because
> n! quickly grows larger than a given random-# generator's period]
>
>[Aahz Maruch]
>> What happens if you re-seed?
>
>By what means?  If you have a good source of random bits to do the reseeding
>with, save yourself the expense of calling random() and just use them
>directly <0.1 wink>.  If your seed generation isn't random, at best you can
>boost the period of the combined generation scheme; at worst, damage the
>pseudo-randomness of both streams.  "It depends", and on everything.

I'm just using the standard time.time() mechanism.  

>> I'm curious because I just wrote some code that re-seeds every
>> hundred calls to random().  In my case, though, it's necessary
>> because I actually don't care that much about randomness per se;
>> I'm using it to generate unique numbers (with duplicate detection),
>> and I need to ensure that if I ever do happen on the same random
>> number sequence as a previous run of the process, it gets aborted
>> reasonably quickly.
>
>Probably better to skip the reseeding, saving random()'s state at
>the end of one run and restoring it at the start of the next --
>random.random()'s period is in the neighborhood of a trillion; it's
>not going to repeat before your current employer goes out of business
><wink>.

That would be a great idea if runs "ended"; however, this is intended to
be a daemon-style process, and it crashes *much* more often than it
ends.  :-(  It certainly wasn't worth the time to create a full-blown
save/restore mechanism for a quick hack.

The reseeding is just a hack on top of a hack as a backstop against the
possibility that time.time() on the start of a new run would duplicate
the seed for a previous run -- I didn't want to see the process spinning
its wheels for however long it took to work through the sequence to the
old ending point.

>But, so far, you haven't said anything that explains why you can't just
>repeatedly add 1 to a global counter!  Guaranteed unique without bothering
>to check for duplicates.  Something else must be driving this.

I don't know what other issues came up, but there's the problem that
currently at least two different processes generate these "serial
numbers".  Could have used the database to store the IDs, of course, but
then you run into the locking problem.

Oh, yeah, I remember another issue.  There's another place where we
expose a different set of serial numbers to the public, and we did *not*
want a way for someone to identify where they were in the sequence.
Until I created this hack, they both used the same serial generation
code.

>> (This is a temporary hack to deal with an unfortunate choice in
>> hashing functions until we can switch to some code that uses SQL
>> Server GUIDs.  And, yes, I did find out about Python's hash()
>> function *after* I wrote this....  <sigh>)
>
>Check out the md5 module for a much stronger (& much slower -- win-win
><wink>)128-bit hash; e.g.,

Yup.  I believe that we initially wanted to use 32-bit ints to make
accessing the database easier; part of the reason we're switching over
is that we didn't realize just how quickly we'd be filling up the 32-bit
address space.  I think I'll suggest to my coworkers that MD5 is more
portable than GUIDs, even though it'll be a bit slower.


Here's another question: I know the *period* of whrandom() is very long,
but what is the granularity/precision of it?  I wasn't sure of the
answer, so I partitioned the 32-bit space into a thousand chunks, making
two calls to random().
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From skip at mojam.com  Tue Aug 24 13:37:26 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue, 24 Aug 1999 12:37:26 -0500 (CDT)
Subject: Pro Python remarks to math ed folks (pointers)
In-Reply-To: <37c2b213.49798667@news.teleport.com>
References: <37c2a51e.18050990@news.teleport.com>
	<Ibjw3.680$v6.61454@typhoon.nyroc.rr.com>
	<37c2b213.49798667@news.teleport.com>
Message-ID: <14274.55328.122770.722670@dolphin.mojam.com>

    Kirby> This 'quad' stuff (with 4-tuples) is the most unusual and exotic
    Kirby> and would stump any classroom math teacher for a few minutes
    Kirby> because it's unfamiliar and not taught as a normal part of any
    Kirby> curriculum.  

Are quadrays similar/identical to quadrics?  If so, why the change in
nomenclature?

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From Wobbler at greenvillenc.com  Fri Aug 20 19:02:56 1999
From: Wobbler at greenvillenc.com (Daryl Zero)
Date: Fri, 20 Aug 1999 19:02:56 -0400
Subject: Thanks for the help. Still left me with a lot of questions but I think I have better direction. What is a troll by the way?
Message-ID: <zelv3.13444$x04.816752@typ11.nn.bcandid.com>

Thanks for the help. Still left me with a lot of questions but I think I
have better direction. What is a troll by the way?





From andyl at azaal.dircon.co.uk  Sun Aug 15 07:31:40 1999
From: andyl at azaal.dircon.co.uk (Andy Leighton)
Date: 15 Aug 1999 12:31:40 +0100
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk>
Message-ID: <slrn7rd98r.upj.andyl@azaal.dircon.co.uk>

Robin Becker <robin at jessikat.demon.co.uk> wrote:
>on steroids and lsd). They probably win on numbers who use it and on the
>module CPAN thing.

The CPAN repository for modules is a win for Perl.  But I also wouldn't
underestimate some of Perl's other nice features like perldoc and MakeMaker.

I must admit I do use both languages (although I am only learning Python),
and I find that when I use Perl it is too easy to fall into bad habits,
and I tend to spend a lengthy session refactoring my Perl programs after 
writing them.

-- 
Andy Leighton => andyl at azaal.dircon.co.uk
"... January is your third most common month for madness" - _Sarah Canary_



From wtanksle at dolphin.openprojects.net  Thu Aug 26 20:12:41 1999
From: wtanksle at dolphin.openprojects.net (William Tanksley)
Date: Fri, 27 Aug 1999 00:12:41 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net> <37C552DF.9F90BA36@megsinet.net>
Message-ID: <slrn7sblvo.p96.wtanksle@dolphin.openprojects.net>

On Thu, 26 Aug 1999 10:44:48 -0400, David Oppenheimer wrote:

>That being said, are there ways to integrate Python and Perl to do things
>that neither one can do alone?!

Minotaur is an attempt to do exactly that -- they've got Python, Perl,
Tcl, Ruby, Forth, and several other languages all glued together and all
callable from each other, with good data interchange.

If anything can result in a synergistic effect, that's it.

>David O.

-- 
-William "Billy" Tanksley



From robert_j_roberts at rl.gov  Sun Aug 29 19:29:36 1999
From: robert_j_roberts at rl.gov (Roberts, Robert J)
Date: Sun, 29 Aug 1999 16:29:36 -0700
Subject: fo.flush() & fo.write() on windows
Message-ID: <37C9C25F.DE0BEEAB@rl.gov>

Am I asking too much?

pseudo code:

    fo = __builtin__.open(foName, 'a')
    foSize = os.stat(foName)[6]     # if foSize = 100...
    fo.write('Some string.')             # fo.write() returns None
    foSize = os.stat(foName)[6]    # ...foSize STILL = 100.
    fo.flush()
    foSize = os.stat(foName)[6]    # ...foSize STILL = 100.
    fo.close()                                # 'Some string' appears in
fo.
    foSize = os.stat(foName)[6]    # foSize = foSize + 'Some string' #
of bytes.

I am running Python 1.5.2 on Windows 95. fo.write() appears to be
buffering the write and returning None for bytes written on fo.write().
fo.flush() appears to do nothing.

...or... am I missing something?





From chad at vision.arc.nasa.gov  Thu Aug 19 21:42:32 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Fri, 20 Aug 1999 01:42:32 +0000
Subject: Unsorting(randomizing) a sequence
References: <000701bee917$d8161cc0$eaa22299@tim> <14268.10663.137315.893101@buffalo.fnal.gov> <ZE1v3.14955$gO1.492954@news2.giganews.com>
Message-ID: <37BCB288.B4476B8B@vision.arc.nasa.gov>

Christopher Browne wrote:

> Is it so much work to come up with a half-reasonable card shuffling
> algorithm?

Well, let's see...

> The one I've always seen cited as providing decent behaviour looks
> thus:
>
> for (i = 0; i > n; i++) {
>    swap (item[i], item[random(n)]);
> }

A list of n items has n! permutations.  The algorithm you propose will
generate  n**n possible execution paths (for lack of a better term), which
is not evenly divisible by n! .  So your variation does not produce all
possible
permutations with equal probability.

The solution posted previously by Dan Schmidt, should work (I'll translate
to C pseudocode to match the above form):

for (i = n; i > 0; i--)
{
  j = i -1;
  swap(item[j], item[random(j)]); /* choose from remaining items in list */
}

The basic difference is that the second approach, after placing
a random item in a slot, only allows the remaining items to
be eligible for the remaining slots.  Think of it as randomly
picking cards from a deck to create a new shuffled deck.

Chad Netzer

PS.  if I f*cked up, be gentle. :-)





From mwh21 at cam.ac.uk  Sat Aug  7 05:03:31 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 07 Aug 1999 10:03:31 +0100
Subject: why are dlopen flags different between 1.5 and 1.5.2?
References: <99Aug6.152924pdt."292365"@ngsun1.parc.xerox.com>
Message-ID: <m3oggkvu70.fsf@atrus.jesus.cam.ac.uk>

Bill Janssen <janssen at parc.xerox.com> writes:

> After a fair amount of grief, I found out that while Python 1.5
> specified RTLD_GLOBAL in its call to dlopen (in Python/importdl.c),
> Python 1.5.2 does not.  Is there any particular reason to change this?

It was thrashed out here a while's back. www.deja.com is your
friend... I think there was also discussion on PSA-members or some sig
or other.

I think the basic gist was that if not using RTLD_GLOBAL causes you
problems you can do something about it, whereas if using RTLD_GLOBAL
causes you problems you are screwed.

> To be compatible with Java?

I doubt it.

> Bill

HTH
Michael



From tre17 at pc142.cosc.canterbury.ac.nz  Mon Aug 30 03:40:26 1999
From: tre17 at pc142.cosc.canterbury.ac.nz (Timothy R Evans)
Date: 30 Aug 1999 19:40:26 +1200
Subject: Question: lists in classes
References: <37CA1803.2FA9261B@ug.cs.su.oz.au>
Message-ID: <c75btbp3edx.fsf@pc142.cosc.canterbury.ac.nz>

Jeff Turner <jturner at ug.cs.su.oz.au> writes:

> Apologies if this has been answered before...
> 
> class foo:
>     list=[]
> 
> a=foo()
> b=foo()
> a.list.append(123)
> print b.list
> 
> I would expect the above code to return "[]", since b has not been
> modified since instantiation. However it returns [123].
> 
> How would I tell python that I only want to _define_ the class, not
> instantiate it?
> 
> Thanks,
> Jeff

The code above adds list as a class attribute, the same list is found
from all instances.  If you want to create an instance, you add it in
the __init__ method as so:

class foo:
    def __init__(self):
        self.list = []

a=foo()
b=foo()
a.list.append(123)
print b.list
[]

--
Tim Evans




From mlh at idt.ntnu.no  Sat Aug 14 18:28:51 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 15 Aug 1999 00:28:51 +0200
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <37b5a8e0@cs.colorado.edu>
Message-ID: <y0jd7wqf13w.fsf@drue.idi.ntnu.no>

Tom Christiansen <tchrist at mox.perl.com> writes:

> 
> As to why there's no x++ there, consider how often people from a
> Pascal(ish) background are confused between x+1 and x++ and ++x, and
> you'll begin to see what kind of problems caused it to be vetoed.
> 

Hm. I would have thought that the reason for this is more that x++ is,
in fact, an expression, an allowing this would amount to allowing a
special case of assignment expressions... Or?

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From pj at sgi.com  Fri Aug 27 21:01:00 1999
From: pj at sgi.com (Paul Jackson)
Date: 28 Aug 1999 01:01:00 GMT
Subject: Wood, metal and plastic [was: Re: Why do we call python scripting?]
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com>
Message-ID: <7q7ccc$5sh44@fido.engr.sgi.com>

Construction materials have two dimensions:

    1) ease of fabrication, and
    2) strength in use.

One constructs a house mostly of wood boards, because it is
cheaper and easier to build with them.  But in critical places
such as locks, hinges and fasteners, one uses metal.

I think of full-blown C++ as die cast metal, and 'classic'
scripting languages as wood.

    Python is plastic.

It's suitable to a wider range of uses than shell scripts or
Perl (though it cannot displace all uses of C or C++) but it
is as easy to work with as scripts.

	wood    == script   [jcl, 4dos, ksh, Perl]
	metal   == compiled [Cobol, Fortran, C, C++]
	plastic == ???	    [Python]

We need a new word - though until Python has a competitor, that
is difficult.  Human brains don't do well picking names for
classes with one known member.  The sound of one hand clapping
or some such.

Or perhaps Python _does_ have plastic competitors.

Are there other "plastic" languages out there that I don't
know about?  If there were, then we could identify the class,
show at least two example members, and 'name that class'.
Then perhaps this 'is Python a scripting language' question
would rest in peace.

(Actually, the way I've heard some of the Lisp crowd speak of
their various dialects and common usage patterns, that sounds
pretty plastic to me.  Python differs in having a syntax
more like the main stream languages, and in not presuming an
elaborate support environment, both of which make Python much
more accessible to a main stream programmer, for main stream
systems, like myself and the systems to which I program.)
-- 

=======================================================================
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj



From gmcm at hypernet.com  Mon Aug 30 09:37:46 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 30 Aug 1999 08:37:46 -0500
Subject: But Sir, It's Wafer Thin!!! 
In-Reply-To: <37CA141D.41285EFD@megsinet.net>
Message-ID: <1276106072-51511585@hypernet.com>

David Oppenheimer writes:

> I only have a small question.  In fact you might say its wafer thin.
> 
> Q: What path declarations do I need to place into my autoexec.bat
> file so that my computer will be able to run Python Scripts?

If properly installed, nothing. Having the path to python.exe there 
is handy.

[tcl / tk]
> ... I used the find command to locate both of these
> dll's and they are here C:\TclPlug\2.0\bin

That's not standard. If that's a real installation, and you want to 
keep it there (and it's the right version - 8.0), first make sure 
that C:\TclPlug\2.0\bin is on your path. Then make sure that you have
C:\TclPlug\2.0\lib\tcl8.0 and C:\TclPlug\2.0\lib\tk8.0 directories 
with .tcl scripts in them. If Python still fails to load Tk, add 
these environment variables to your autoexec.bat:

TCL_LIBRARY=C:\TclPlug\2.0\lib\tcl8.0
TK_LIBRARY=C:\TclPlug\2.0\lib\tk8.0


- Gordon



From dek at nano.ucsf.edu  Sat Aug 14 20:53:48 1999
From: dek at nano.ucsf.edu (David Konerding)
Date: 15 Aug 1999 00:53:48 GMT
Subject: Anyone else find Tk akward?
References: <37B34DBB.E5A79760@sprint.ca>
Message-ID: <slrn7rc3ss.88i.dek@nano.nmr.ucsf.edu>

On Thu, 12 Aug 1999 15:42:03 -0700, Orlando Vazquez <ovazquez at sprint.ca> wrote:
>Is it just me, or is the way Tk organizes things on the window someone
>weird? I'm just starting to play around with it, and it has me puzzled.
>Wouldnt it be a lot easier to just use a coordinte system?
>For example put a button at (0,0) with length 10 and width 20 etc..
>Are there any reasons why it _has_ to be this way?

There are several geometry managers for Tk.  There's place, pack, and grid.
I liked grid because it gave me direct control over subwidget placement.
Since you can nest frames with different geometry managers, for example,
a grid frame with pack frames inside of it,  you've got the best of both
worlds.

Dave
-- 
--------------------------------------------------------------------------------
Email: dek at cgl.ucsf.edu    David Konerding     WWW: http://picasso.ucsf.edu/~dek
--------------------------------------------------------------------------------
Snail: Graduate Group in Biophysics
Medical Sciences 926, Box 0446
University of California
San Francisco, CA 94143



From possum at n-space.org  Thu Aug 12 17:54:32 1999
From: possum at n-space.org (Erik Stewart)
Date: Thu, 12 Aug 1999 21:54:32 GMT
Subject: Help!  $PYTHONPATH trouble
References: <37B33980.A5318181@n-space.org> <n5Hs3.14$2y4.6929@news-dal.corridex.com>
Message-ID: <37B3424A.6468237A@n-space.org>

Evan Simpson wrote:
> 
> First, you need to either export PYTHONPATH or use the assignment as a
> prefix, thus (assuming bash):
> 
> $ PYTHONPATH="home/possum/test/kwp"
> $ export PYTHONPATH
> $ python
> or
> $ PYTHONPATH="home/possum/test/kwp" python

Thanks!  It works fine now!  I knew I must be missing something simple,
but I couldn't figure out what.

> 
> Second, shouldn't that path start with a "/"?

Oops :@


> 
> Erik Stewart <possum at n-space.org> wrote in message
> news:37B33980.A5318181 at n-space.org...
> The variable $PYTHONPATH doesn't seem to have any effect on my system (RH
> Linux 6.0, Python 1.5.1).  Here's what I get after defining $PYTHONPATH:



From bowman at montana.com  Sat Aug 21 13:43:39 1999
From: bowman at montana.com (bowman)
Date: Sat, 21 Aug 1999 11:43:39 -0600
Subject: Book -- or website -- about Tk & tkinter
References: <935172552snz@vision25.demon.co.uk> <37BE30BA.E45BB293@montana.com> <935226988snz@vision25.demon.co.uk>
Message-ID: <37BEE54B.C517B121@montana.com>

Phil Hunt wrote:
> 
> Is the Lif Preserver still useful? Lundh's _An Introduction to
> Tkinter_ says it's a replacement for it.

I use a local copy of LifePreserver as a quick reference, and have a
Netscape button aimed at it. I have the Intro as a ps file, but
ghostview doesn't really make my day. 

I'm sure Lundh work will be a replacement someday, but that is then...
Right now, it seems the areas I'm interested in are largely TODOs.
Menus, for instance.

> But you're right -- experimentation will no doubt be necessary.

I'm mostly faking it, working from Tcl/Tk, PerlTk, Motif, X docs, and
prior experience with Java and MS Windows programming in C. Essentially,
I'm trying to develop a Tkinter style that is consistent with what I do
in other languages. This sometimes leads me to interesting (to me)
tricks. I'm used to using pretty much anonymous widgets, and determining
their identity in the callbacks, leading to hacks like:

   for b in ('one', 'two', 'three'):
	b = Button(parent, text=b)
	b['command'] = lambda wid=b : (someFunc(wid))
	b.pack(side=LEFT)

Its real easy to whip up a Tkinter GUI for a specific set of widgets,
which is what most of the examples do. Trying to use another level of
generalization, say if you want to generate a GUI from something like a
uil or rc file, or in your case, Parrot, requires a little more work.

> Ah yes, Java's GridBagLayout widget.

Yeah, wondrous thing, that. In Java, I tend to set the layout manager to
null and do absolute positioning. Or, I should say, I calculate the
positions and then place things, effectively writing my own manager.
Moot point. It gets ugly, especially with text, where you need to get
the TextMetrics, or whatever the local Java/Windows/Xt word for it is.
There is no free lunch, though, and someplace, somewhere, somebody is
doing the math. At least if you do it yourself, there are no surprises
-- or you can fix your own little surprises.

You might want to take a look at place(), too. The relative positioning
isn't real intuitive, but it sometimes helps to place a widget at 33% of
the width of the parent, no matter what happens to the parent.


-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From monica at amateuruk.com  Fri Aug 27 22:56:28 1999
From: monica at amateuruk.com (Monica James)
Date: 28 Aug 1999 02:56:28 GMT
Subject: AMATEUR GIRLS ARE HERE XXX
Message-ID: <7q7j4s$ipu$4837@nclient11-gui.server.virgin.net>

ALL NEW ALL AMATEUR SITE XXX

http://212.67.128.13/home/tool



From larsga at ifi.uio.no  Sun Aug  1 17:24:14 1999
From: larsga at ifi.uio.no (Lars Marius Garshol)
Date: 01 Aug 1999 23:24:14 +0200
Subject: RMI and Python
References: <Pine.GSO.4.10.9908011131490.25758-100000@xmission.xmission.com>
Message-ID: <wkhfmj6vpd.fsf@ifi.uio.no>

* Tim Egbert
|
| Java is able to do remote method invocation (RMI) which is a neat
| concept.  It is the basis for Jini, another neat idea.  Does Python
| have anything like RMI?

Yes: CORBA. See <URL: http://www.dstc.edu.au/Fnorb/ >. I think there
also are some bindings to various C ORBs floating around. Compared to
RMI CORBA has the advantage that it is completely vendor-, language-
and platform-independent. 

--Lars M.



From mal at lemburg.com  Sat Aug 14 17:13:04 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sat, 14 Aug 1999 23:13:04 +0200
Subject: Mac ODBC
References: <7p4f3r$16t2@dilbert.whoi.edu>
Message-ID: <37B5DBE0.31F33C@lemburg.com>

Dan Grassi wrote:
> 
> Does anyone have ODBC working under Mac Python?  If so could you provide me
> with some hints on where to get the various Mac sources and some compile
> hints?  I do have a PPC Mac ODBC driver that does work.

Does the driver expose the ODBC API in a way that allows you to
link against it ? If yes, then you could try to create a new
subpackage for mxODBC (see my Python Pages for more infos).

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   139 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From frehe491 at student.liu.se  Mon Aug 23 10:56:35 1999
From: frehe491 at student.liu.se (Fredrik Henbjork)
Date: Mon, 23 Aug 1999 16:56:35 +0200
Subject: Why use Perl when we've got Python?!
References: <slrn7s2lb3.k6b.abigail@alexandra.delanet.com>
Message-ID: <99082317011801.20344@o112>

On Mon, 23 Aug 1999, you wrote:
> Larry and Tom can teach all they want; their teachings will have no impact
> on Computer Science what so ever. Just like their teaching will have no
> impact on the grow of potatoes in Idaho.

Who are you who are so wise in the ways of science?
I am Larry, King of the Perl Mongers.
My liege!
Good Sir knight, will you come with me to Usenet, and join us at c.l.p.m?
My liege! I would be honored.
What is your name?
Tom, my leige.
Then I dub you Sir Tom, Knight of c.l.p.m.
<piwink>

--
Fredrik Henbjork

Support freedom of mind. Creative thoughts for a better tomorrow!
http://www.maps.org | http://www.lycaeum.org | http://www.norml.org



From clgonsal at keeshah.penguinpowered.com  Mon Aug 23 16:08:47 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Mon, 23 Aug 1999 20:08:47 GMT
Subject: Q: Authentication with httplib
References: <7pr09c$dag$1@news.netvision.net.il>
Message-ID: <slrn7s3ai4.rc.clgonsal@keeshah.penguinpowered.com>

On Mon, 23 Aug 1999 11:27:22 +0200, Zwika Moshkowzki <zwi at turboimage.com> wrote:
> Hi,
> 
> I am using httplib to access a Page on a Server:
> 
>     .....
>      h = httplib.HTTP(Server)
>      h.putrequest('GET', Page)
>      h.endheaders()
>     .....
> 
> When Server is configured not to require a passwd, I set Server="myServer",
> Page="/myPage",
> and everything is OK.
> 
> My Q regards the situation where there is a need for authentication: Where
> should I put the Username+Passwd?
> 
> Thanks in advance!

The username and password need to go in an "Authorization" header.
Something like this should work:

    import base64
    import string
    import httplib

    userid = "foo"
    passwd = "bar"

    # "Basic" authentication encodes userid:password in base64. Note
    # that base64.encodestring adds some extra newlines/carriage-returns
    # to the end of the result. string.strip is a simple way to remove
    # these characters.
    auth = 'Basic ' + string.strip(base64.encodestring(userid + ':' + passwd))

    h = httplib.HTTP(Server)
    h.putrequest('GET', Page)
    h.putheader('Authorization', auth )
    h.endheaders()

Hope that helps...

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From fredrik at pythonware.com  Fri Aug 27 04:55:27 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Fri, 27 Aug 1999 08:55:27 GMT
Subject: Library philosophy
References: <37C57A80.D5C62704@prescod.net>
Message-ID: <02a101bef069$f1ecc190$f29b12c2@secret.pythonware.com>

Paul Prescod <paul at prescod.net> wrote:
> Would it it make sense to start wrapping the low-level modules with
> high-level, easier to use modules that would be "the default" in Python
> 2?

assuming that Python 2 is what will come out of
the CP4E project (*), that's a must.

*) http://www.python.org/doc/essays/cp4e.html

on the other hand, it wouldn't surprise me if
Barry and Fred has already cleaned up the 1.6
library, while Guido was out of town... ;-)

</F>




From tchrist at mox.perl.com  Sat Aug 14 22:39:11 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 14 Aug 1999 19:39:11 -0700
Subject: Why use comp.lang.python when you've got comp.lang.perl.misc?
References: <37B5CB95.3FCF571F@appliedbiometrics.com> <199908142005.OAA23215@jhereg.perl.com> <m27lmyj8nv.fsf@netmeg.net>
Message-ID: <37b61a3f@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Les Schaffer <godzilla at netmeg.net> writes:
:hey, why should the well behaved among us have to wait for
:flame-throwers to chill out. we're here to learn about the languages
:we use. bring on your insights, brother.

It'll take me a little while to type them all in.  I have pages and
pages of longhand notes, yet to be distilled.  I'm not sure that I'll
be done in time for the Open Source conference.  It's intended as a
non-judgmental set of corresponding features/properties/attributes.

--tom
-- 
    #else /* !STDSTDIO */     /* The big, slow, and stupid way */
        --Larry Wall in str.c from the 4.0 perl source code



From miker3 at ix.netcom.com  Sun Aug 15 08:09:57 1999
From: miker3 at ix.netcom.com (Michael Rubenstein)
Date: Sun, 15 Aug 1999 12:09:57 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu>
Message-ID: <37b9a99b.635832158@nntp.ix.netcom.com>

On 14 Aug 1999 21:31:46 -0700, Tom Christiansen
<tchrist at mox.perl.com> wrote:

>Think back to the 80s and early 90s.  The change, if it really happened,
>came with the upty-gazillion CGI script kiddies who couldn't program
>2+2, yet who wanted *us* to write *them* these persistent, encrypted,
>crossplatform, multiscreen shopping carts for e-commerce, complete with
>dynamically-generated animated vanity counters, each and every time 
>they stepped up to the feeder.
>
>As a result, There's not much left of a spirit of a programming community
>of mutual benefit when it becomes numerically dominated by people who
>don't want to study or to work or EVEN TO PROGRAM, and who all seem to
>think you owe them something.

Anyone in comp.lang.python who is starting to feel sorry for the
beleaguered perl experts should read comp.lang.perl.misc.

Don't miss the thread (checking Perl offline) in which Tom goes
after one of these "CGI script kiddies" who has the audacity to
use the word "off-line" to mean not on the network and "upload"
to mean copy to the server (both usages are correct according to
the OED).

Make sure you check out the spelling and grammar from the
regulars.  Many of them are fond of insulting beginners for not
reading news.announce.newusers but feel free to ignore what the
announcements say about spelling and grammar flames.  Make sure
you check out the thread (Nastiness contrary to the spirit of
perl?) in which one of the regulars feels it necessary to attack
someone for using the word "ditto" to mean "I agree" (a
reasonable use of the word according to the OED).

Nor should you miss the current thread in which a number of
regulars register their reaction to Tom's harrassing someone with
email (HARASSMENT -- Monthly Autoemail) because Tom doesn't
approve of his method of quoting.

The problem in comp.lang.perl.misc is not the beginners -- it is
the regulars.



From tchrist at mox.perl.com  Sun Aug 15 00:31:46 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 14 Aug 1999 21:31:46 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com>
Message-ID: <37b634a2@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.perl.misc, thehaas at my-deja.com writes:
:Although Perl and Python, as languages, are very similar, the culture
:that surrounds the two languages are quite different.  I know when I've
:asked someone for help in either Python or Perl, the Python people are
:definitely the most helpful.  I think I can safely say, that the Perl
:community is quite arrogant, especially to people new to it.  

It certainly didn't start out that way, and for many years Perl was
famous for being precisely exactly the opposite of hostile.  It was
about a programming community of mutual benefit.

Think back to the 80s and early 90s.  The change, if it really happened,
came with the upty-gazillion CGI script kiddies who couldn't program
2+2, yet who wanted *us* to write *them* these persistent, encrypted,
crossplatform, multiscreen shopping carts for e-commerce, complete with
dynamically-generated animated vanity counters, each and every time 
they stepped up to the feeder.

As a result, There's not much left of a spirit of a programming community
of mutual benefit when it becomes numerically dominated by people who
don't want to study or to work or EVEN TO PROGRAM, and who all seem to
think you owe them something.

When > 90% of the newsgroup postings are ill-formatted garbledy-gook
that also fall into the category of not having bothered to have first
checked the online docs or faqs that come with Perl, these things add up.
And eventually, it risks breaking our spirit.  There's just no way to deal
with the never-ending onslaught of non-programmer CGI script kiddies, who
seem to outnumber the rest of us zillions to one.  

So what do you do?  You either bail out of this untenable situation, as
Larry Wall has done, or else "customer-service battle fatigue" sets in.
When that happens, people end up getting snapped at, sometimes mistaking
an honest learner for yet another beggar looking for handouts.  I dare
say this would happen even in the best of families, given equivalent
circumstances.  I do think that earnest individuals seriously wanting
to work hard are still given the same helping hand they always got.

Be careful what you wish for.

--tom
-- 
Emacs is a fine operating system, but I still prefer Unix. -me



From ovidiu at cup.hp.com  Fri Aug 13 16:42:33 1999
From: ovidiu at cup.hp.com (ovidiu at cup.hp.com)
Date: Fri, 13 Aug 1999 13:42:33 -0700
Subject: Putting code somewhere 
In-Reply-To: Your message of "Fri, 13 Aug 1999 20:26:40 GMT."
             <4c%s3.1113$hv3.57169@typhoon1.gnilink.net> 
Message-ID: <199908132042.NAA25893@hercules.cup.hp.com>

On Fri, 13 Aug 1999 20:26:40 GMT, "Damond Walker" <dwalker at iximd.com> wrote:

>     I'm wrapping up a simple curses-based interface module (implements
> textboxes, buttons, and labels -- told ya it was simple).  It is to be used
> for console applications and supports a notion of dialog boxes as well.
> Assuming that anyone has any interest in this, where would I go about
> posting the code?  python.org?

I don't know exactly what you've built, but there is a C library out there that
does the same thing. Check out Curses Development Kit (CDK) at:

http://www.vexus.ca/CDK.html

The list of implemented widgets is impressive, check it out. Would be nice to
have a Python binding to CDK, like Perl has.

Best regards,
--
Ovidiu Predescu <ovidiu at cup.hp.com>
http://andromeda.cup.hp.com/  (inside HP's firewall only)
http://www.geocities.com/SiliconValley/Monitor/7464/






From a.mueller at icrf.icnet.uk  Thu Aug 12 12:11:38 1999
From: a.mueller at icrf.icnet.uk (Arne Mueller)
Date: Thu, 12 Aug 1999 17:11:38 +0100
Subject: MySQL module crashes
Message-ID: <37B2F23A.7E084E5C@icrf.icnet.uk>

Hi All,

I've just installed MySQL 1.4, mysql 3.22.25 and python 1.5.2b2

Unfortunately python coredumps :-(

mueller at jumbo:~/src/Python-1.5.2b2> python
Python 1.5.2b2 (#17, Aug 12 1999, 16:53:47) [C] on irix6
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
python> import MySQL
Fatal Python error: PyThreadState_Get: no current thread
Abort (core dumped)

I'm runniing irix 6.5 on R5000/10000. Does anybody have any 
experience with mysql module in the above version combination?
In general I've no problems with threads in python.

	thanks for help,

	Arne



From esuzm at primrose.csv.warwick.ac.uk  Tue Aug 17 09:49:32 1999
From: esuzm at primrose.csv.warwick.ac.uk (Andrew Clover)
Date: 17 Aug 1999 13:49:32 GMT
Subject: Bug in ntpath.exists?
Message-ID: <7pbp9c$1b4$1@holly.csv.warwick.ac.uk>

This is odd.

There seems to be a problem in os.path.exists on WinNT, and presumably the
other os.stat-based calls in ntpath.

It returns true (well, 1) for objects with any number of spurious '.' and '"'
characters on the end of a filepath, and any number of spurious spaces on the
end of the filepath after these. Additionally any path element may contain a
single spurious '.' character at the end. (Also each path separator may
consist of any number of forward- or back-slashes, but that's probably
intended.)

That is, if the file 'e:\fish\chips.txt' exists, the following paths are also
claimed by exists to exist:

  'e:\fish\chips.txt.'
  'e:\fish.\chips.txt"""'
  'e:\fish/\/\/\chips.txt....      '

and so on.

This leniency could be considered a good thing in some cases, but not in my
case since I'm writing a 404 CGI script which attempts to verify the missing
URL searched for and corrent wrong parts.

Actually open() will deal with most of these 'wrong' paths and open them
anyway, except for the case where a spurious '"' character is on the end of
a filename.

Should I fix this by bodging ntpath, or should something be changed in
os.stat, or what?

-- 
This posting was brought to you by And Clover.
(Sorry.)



From jeffp at crusoe.net  Sun Aug 15 17:18:39 1999
From: jeffp at crusoe.net (Jeff Pinyan)
Date: Sun, 15 Aug 1999 17:18:39 -0400
Subject: C's isprint() concept?
In-Reply-To: <slrn7rea88.9uf.jblaine@shell2.shore.net>
References: <slrn7rea88.9uf.jblaine@shell2.shore.net>
Message-ID: <Pine.GSO.4.10.9908151713380.3761-100000@crusoe.crusoe.net>

> If I want to replace all non-printable characters in a string with a single
> space, what would be the best way?  Do I need to loop over the entire string
> character by character checking the ord() value of each one?  Anyone have
> a sane way to do this with regular expressions?

In Perl, I could do it this way:
  $string =~ tr/\x00-\x1f\x80-\xff//d;

What that means is this:
  remove the following characters from $string:
    characters whose ASCII value is from \x00 (0) to \x1f (31)
    characters whose ASCII value is from \x80 (128) to \xff (255)

This can be done, almost as painlessly, in Python.

--
jeff pinyan    japhy at pobox.com  japhy+perl at pobox.com  japhy+crap at pobox.com
japhy's little hole in the (fire) wall:       http://www.pobox.com/~japhy/
japhy's perl supposit^Wrepository:       http://www.pobox.com/~japhy/perl/
The "CRAP" Project:                 http://www.pobox.com/~japhy/perl/crap/
CPAN ID: PINYAN           http://www.perl.com/CPAN/authors/id/P/PI/PINYAN/




From m.faassen at vet.uu.nl  Mon Aug  9 04:59:32 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 9 Aug 1999 08:59:32 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <19990805094020.A1838753@vislab.epa.gov> <37aacdf9.324492916@news.oh.verio.com> <934117874snz@vision25.demon.co.uk>
Message-ID: <7om59k$cqc$1@newshost.accu.uu.nl>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
> I'm currently writing a generic GUI builder.

> It doesn't (and won't) have a graphical user interface, the user
> will write a user interface definition file, which will then be
> compiled by my tool into the programming language and GUI toolkit of
> your choice. A definition file might look like this:

[snip definition file example]

Why not use XML for this?

[snip]
> The idea is that it will, in principle, be able to generate code for 
> any back end. (Once someone has written that back end, of course).
> The advantage of this is that the UI definition language only needs 
> to be learned once, and can be used for lots of languages and GUI
> toolkits (of which there are a lot around).

I like this idea; I wish you luck!

Regards,

Martijn




From I.Clarke at NOSPAM.strs.co.uk  Thu Aug  5 04:44:10 1999
From: I.Clarke at NOSPAM.strs.co.uk (Ian Clarke)
Date: Thu, 05 Aug 1999 09:44:10 +0100
Subject: Hidden CGI fields in HTML
Message-ID: <37A94EDA.81644EF3@NOSPAM.strs.co.uk>

Hi,

I am using CGI with Python and am trying to find a way that I can
generate the HTML for a form such that one of the INPUT controls will
have a fixed value (set when the page is generated by a previous CGI
script), that CANNOT be edited by the user (or prefereably not even
seen) - but will be passed to the next CGI script.  

I am aware that there are "readonly" and "disabled" tags for INPUT
controls defined in the HTML 4 spec, and even a style="display:none"
attribute should work in theory, but none of these seem to work with
Netscape.  The readonly and disabled tags have no obvious effect, and
while setting the style to "display:none" does stop it being displayed,
it also stops the parameter being passed in the form!

Has anyone actually achieved this seemingly simple task?

Cheers,

Ian.
-- 
Remove the "NOSPAM." from my email address before replying.



From mstenber at cc.Helsinki.FI  Wed Aug  4 12:22:18 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 04 Aug 1999 19:22:18 +0300
Subject: clues to source of this delay?
References: <al8zp07pwyd.fsf_-_@sirppi.helsinki.fi>
Message-ID: <al8n1w7plc5.fsf@sirppi.helsinki.fi>

Markus Stenberg <mstenber at cc.Helsinki.FI> writes:
> Markus Stenberg <mstenber at cc.Helsinki.FI> writes:
> > And the CPU spent is about 2%, thus the delays are .. somewhere. Any
> > insights on where? The Linux performance is sufficient, but NT performance
> > most definitely is _NOT_.
> To follow-up to my own query:
> 	- NT4 server did _not_ perform any better (5 roundtrips/sec, @whee)
> 	- Eliminating select altogether did not help at all
> Thus, it seems it's out of Python's hands for 95% of the
> time. .. sigh. This twilight zone stuff just isn't for me :P

.. final update (hopefully):

Just for laughs I wrote the simplified version of the whole thing in C;
basically, a server that accepts only one connection in recv+send+send
loop, client with connect and then send+send+recv loop. Timings did not
change at all.

Problem was _still_ unsolved (something's out there, mr. Mulder).  Values,
respectively: NT 2.5, HPUX 3.5, Linux 25 (dual roundtrip's) per second.
For some reason, offers of 'get a Linux box' will probably receive great
deal of scorn; thus I'd like to see this solved, one day.. siigh. :-P

And as if to scorn me, the _simplistic_ send+recv clients written in Python
continue to score much faster roundtrips, despite the fact that the
packets' _length_ is the same, even. Grrr....

Finally I observed the problem in my protocol; number of sends per
iteration != number of recvs per iteration (per participant).  In general
(for each received non-ack message, the client/server sends ack _and_ new
message in the echo mode). For some reason, TCP's slow start on the boxen
involved did _not_ work as advertised, and therefore window size stayed at
1 and the speed stayed constant.

That's what I figured, anyhow; after a small rewrite the new protocol with
slightly better buffering, and therefore just send+recv/recv+send cycles,
gets fairly close to saturating my ethernet.

-Markus, signing off after a fairly frustrating 13+-hour code marathon.

-- 
----- BEGIN GEEK CODE v3.0 BLOCK (*) -----
GCS d- s+: !a C+(++) UL+++$ P--- L+++ E++ W-- N+ W--- O---- M-- V- PS+++
Y+ PGP++>+++ t+@ 5++ X+ R* tv-- b+++ DI+++>++++ D---- G++ e++ h* r% z+
----- END GEEK CODE v3.0 BLOCK -----
* For spoilers, check http://www.geekcode.com



From jim at interet.com  Mon Aug 16 09:28:50 1999
From: jim at interet.com (James C. Ahlstrom)
Date: Mon, 16 Aug 1999 09:28:50 -0400
Subject: 16-bit Python -- does it exist?
References: <MPG.122214458f7ab0d2989685@news.mop.no>
Message-ID: <37B81212.733CDD9B@interet.com>

Alexander Staubo wrote:
> 
> Subject says all. A friend (really!) would like to run Python on an old
> 80286. Possible?

Python for 32-bit DOS and for Windows 3.1 are on the
download page www.python.org/download.

Jim Ahlstrom



From thomas.heller at ion-tof.com  Thu Aug 19 05:20:38 1999
From: thomas.heller at ion-tof.com (Thomas Heller)
Date: Thu, 19 Aug 1999 11:20:38 +0200
Subject: Win32 Installer - new beta
References: <1278333954-16187909@hypernet.com> <37BB1373.6686F844@aw.sgi.com>
Message-ID: <00c501beea24$19f87c70$4500a8c0@thomasnotebook>

[description of Gordon's win32 installer deleted...]
> There appears to be two differences between Run.exe and Runw.exe:
>  (1)  Run.exe is a console app, and Runw.exe is not.
>  (2)  Run.exe cleans up all its extracted files, and Runw.exe does not.
[...]
>
> As I said before, I really like this package.  If I could get rid of the
> console I'd REALLY like this package.  I may attempt to compile my own
> version of Run?.exe, but I'm an old Unix weenie still learning to swim
> in the deep end of the Windows pool.  Any help you could provide would
> be appreciated.
>
> Thanks
> Gary Herron

Gary,

I have added new features to Gordons installer package and did send him
the code to include in his distribution.
He has not yet made them available, so I uploaded them to my starship page.

The executables are compiled with the cleanup flag, so they may?be what
you need. Also included is a VC6.0 workspace so you can play with the flags.

You may also be interested in my improvements: It is now possible
to build executables which only need python15.dll, zlib.pyd and
exceptions.pyc.
Everything else is contained in the exe file and never needs to be unpacked.

The URL is http://starship.python.net:9673/crew/theller/installer.

Regards

Thomas Heller




From tim_one at email.msn.com  Fri Aug 20 05:39:30 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Fri, 20 Aug 1999 05:39:30 -0400
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer) 
In-Reply-To: <199908200704.BAA18741@jhereg.perl.com>
Message-ID: <001601beeaef$e795f380$712d2399@tim>

[posted & mailed]

[tchrist, discovers that Python "def" is executable, and notes that
 this is a gotcha for Perl programmers]

[Tim, notes that Perl's "sub" is equally a gotcha for Python programmers]

[back to tchrist]
> You didn't localize your function!  Remember, in Perl things
> are only local if you say they are.

But it doesn't matter, Tom <0.5 wink>:  the "gotchas" you identified in
Python are things that would throw a Perl programmer if they hadn't read and
understood the Python docs first.  Perl is at least equally as surprising to
the Python programmer who hasn't read and understood the Perl docs first (as
the self-proclaimed Pythoneers who flame on c.l.p.m. prove more than often
enough).

[and where I had
    sub inner sub { ... },
 do instead]
> ...
> That could have been like this:
>
>     sub outer1 {
>         my $x = 50;
>         local *inner = sub {
>             print "in the first inner, and here's x: $x\n";
>         };
>         &inner();
>    }

No problem, but that's (symbol table globbing) a *real* stretch for someone
just learning the language.  There are real barriers for native speakers of
either language in picking up the other.

I ended up doing something like this instead:

sub makefac {
    my $fac;
    $fac = sub {
        my $n = shift;
        return 1 if $n < 2;
        return $n * &$fac($n - 1);
    };
    return $fac;
}

my $i;
for ($i = 0; $i < 1000000; ++$i) {
    my $fac = makefac;
    my $n = &$fac(3);
    print "$n ";
}

This one is more interesting than it looks <wink>:  in the last version of
Perl I tried it on, it leaked memory at a prodigious rate.  The reason it's
interesting is because it's faking pretty much what Python would have to do
to support lexical closures, given that a nested "def" acts exactly like an
assignment statement binding a name local to its containing function (and
here an anonymous sub is assigned to a lexical local of *its* containing
function).

In Python and in Perl (at least in the Perl I last tried that under) that
creates cyclic garbage:  makefac's lexicals have to contain a binding for
fac, and fac has to point back to makefac's lexicals in order to resolve
up-level references.  And that's (unrecoverable trash) the reason Python
doesn't have lexical closures today.  If a nested def were-- as my original
nested "sub inner" is in Perl --treated as global, that problem would go
away, and for the same reason it does in Perl.  But it wouldn't be Python
anymore either <wink>.

> ...
> BTW, you should have gotten a bucket of frightened warnings out of
> the compiler about that.

I got strong indications of trouble with -w, yes.

[examples of great msgs produced by
     perl -wc
 and
     perl -Mdiagnostics -wc
]

> ...
> BTW, see what I mean about the importance of good error messages?

No argument from me that Perl's msgs are often very helpful, much more often
pinpointing a problem than Python's (the "runaway multi-line string starting
at line nnn" is one Python should steal!).  On the other hand, I've written
a few Klines of production Perl too, and my experience is that Perl has so
many more syntactic and semantic gotchas than Python that the language would
be nearly unusable without great msgs.  Python's NameError may drive you
nuts, but it's the same as Perl's runtime "use of uninitialized value" --
and Perl is no more help in tracking down the *cause* of those than is
Python in cracking a NameError.  At least Perl's -w warns about names that
appear only once!  You have to grab a copy of PyLint for that (but then you
are a fan of separate tools, right <wink>?).

[various ways to promote warning into errors, "if you feel all
 warnings should kill you"]

Whereas Python has no warnings of any kind -- it it's irked enough at me to
complain at all, it's irked enough to kill me.  In that it's a faithful
reflection of Guido's personality <wink>.  Seriously, I do prefer that.

> ...
> You know, it occurs to me that it might be nice to be able to
> selectively promote only compile-time warnings to fatals.

As a native speaker of Python, -w and "use strict" decorate all my Perl, and
in  the few Official Perl projects I've been on "NO WARNINGS!" was an
absolute rule.  The more fatals the better.

> Perl does a lot of compile-time stuff (including constant
> expression folding, inlining of some subroutines, and folding
> of those results!).
>
> What does Python do at compile-time?

It's too busy desperately scrounging for excuses to kill your program with a
bogus SyntaxError to do anything useful <wink>.  About the only compile-time
optimization is that function locals (in the absence of "exec" and "import
*" statements) are mapped to consecutive indices into a contiguous runtime
vector.

"exec" and "import *" can alter the *set* of local names, so their presence
inhibits that optimization.  That's why "exec" is a stmt rather than a
function:  since no builtin names are resolved at compile-time, and any name
may be bound to any object at runtime, if exec were a function Python
couldn't know that

    exec(s)

does refer to the builtin "exec", or that

    f(s)

*doesn't*, so the local-var optimization would be crippled almost always.
So "exec" is a keyword, the only kind of word whose meaning is known for
sure at compile-time.

Python's default arguments are sometimes used to fake lexical closures, but
they're also used to fake the effect of compile-time binding; e.g., because
def *is* executable, when Python (at runtime) gets to

def somefunc(x, int=int):
    ... = int(...)

the local name "int" gets bound to whatever-the-heck the global name "int"
is bound to at the time the def is executed.  Of course it *is* bound to the
builtin int function, but Python can't know that at compile-time.  This
trick lets the programmer force the issue, and within somefunc the "int"
lookup is very fast (just a C indexed array ref); without this trick each
dynamic instance of "int" in somefunc does a failing lookup in the global
dict first, and then a successful lookup in the builtin dict.

So, no matter what anyone here tells you, Perl doesn't have a monopoly on
being cute & repulsive at the same time <0.8 wink>.

takes-one-to-know-one-ly y'rs  - tim






From tre17 at student.canterbury.ac.nz  Tue Aug 31 04:45:53 1999
From: tre17 at student.canterbury.ac.nz (tre17 at student.canterbury.ac.nz)
Date: 31 Aug 1999 20:45:53 +1200
Subject: Bug in Tkinter?
Message-ID: <c75zoz849tq.fsf@pc142.cosc.canterbury.ac.nz>

Have a look at the following code.  It should produce a window
containing two radio buttons, one of which is active when the program
starts.  But it does not!  Now remove the comments, so that there is a 
permanent reference to StringVar instance, and now it works?

This is running Python 1.5.2 from the Debian package `python-tk 1.5.2-5'
under Debian 2.2 (potato).

=====================
#!/usr/bin/python

from Tkinter import *
import sys

class Foo(Frame):
    def __init__(self, master):
        Frame.__init__(self, master, relief='raised', bd=1)
        mode = StringVar()
        self.b1 = Radiobutton(self, text='Foo',
                              variable=mode, value='foo',
                              command=self.foo)
        self.b2 = Radiobutton(self, text='Bar',
                              variable=mode, value='bar',
                              command=self.bar)
        self.b1.pack(side='left', fill='x', expand=1)
        self.b2.pack(side='right', fill='x', expand=1)
        mode.set('bar')
        print sys.getrefcount(mode)
        #d['mode'] = mode
        #print sys.getrefcount(mode)

    def foo(self):
        print 'foo'

    def bar(self):
        print 'bar'

root = Tk()
d = {}
f = Foo(root)
f.pack()
root.mainloop()
=====================

Any comments?

--
Tim Evans



From aahz at netcom.com  Thu Aug 19 14:40:38 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 19 Aug 1999 18:40:38 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <19990819174800.D5227@xs4all.nl> <37bc313b@cs.colorado.edu>
Message-ID: <7phj36$odu@dfw-ixnews8.ix.netcom.com>

[please don't cc me if you follow up]

In article <37bc313b at cs.colorado.edu>,
Tom Christiansen  <tchrist at mox.perl.com> wrote:
>In comp.lang.python, 
>    Thomas Wouters <thomas at xs4all.nl> writes:
>:
>:Only if you aren't used to exceptions. Exceptions are really very simple and
>:nice things. Especially back when they were just strings (before my time, but
>:still ;-)
>
>No, I'm sorry, the expressivity of "break LABEL" and "continue LABEL"
>are much clearer than the cicumlocutions necessary for exceptions.
>That's like saying you can get the same effect by adding a bunch of lines
>of code.  That's not the same as built-in support, nor half so clear,
>nor half so convenient.  Isn't python about rapid prototyping and all?
>If it were about purity over convenience, I'm sure it would have been
>a lot different -- and completely unused.

OTOH, you can raise an exception and an arbitrary bit of code higher up
in the chain can catch it.  Which has good points and bad points; I
certainly won't claim that it discourages obscure code.

As a former perler, I would not say that Python is "about" rapid
prototyping; for me, even a "quick hack" in Python takes more planning
than the equivalent bit in Perl.  However, I do believe that Python
overall is *much* better suited for software engineering; I simply can't
imagine doing a multi-person project in Perl.

You neglected to mention IDLE and the command-line interpreter as a
Coolness, BTW.

>:> GOTCHA: (high)
>:>     Python has no manpages!  The horror!!!!!!!!!!!!!!!!!!!!!!
>:>     ENODOC
>:
>:Damn, i hadn't even noticed that. Can you imagine ? I guess the '__doc__'
>:attributes on objects are enough manpage for me. ;)
>
>It's unconscionable.   Lack of tools and the tool-based 
>approach strikes again.  See other postings.

That's one approach.  HTML, OTOH, leads to a navigation-based approach.
With Lynx as my primary browser, I feel that I get the best of both
worlds.

>:> QUESTION:
>:>     What is and what is not thread safe?
>:
>:Everything is thread safe. Or rather, there will at all times be only one
>:thread executing the Python interpreter.

This is incorrect.  See the docs for whrandom.

>Huh, no threads in python?

Yes, there are threads, but there is a global interpreter lock.
Anything that executes C code releases the lock (in general), so you get
significant parallelism if you're I/O heavy (for example), but not so
much if you're computationally intensive (in Python-only code).
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From jae at ilk.de  Thu Aug 26 06:08:37 1999
From: jae at ilk.de (Juergen A. Erhard)
Date: Thu, 26 Aug 1999 12:08:37 +0200
Subject: Python / Cool looking applications?
In-Reply-To: <37C16469.687EDE58@am.appstate.edu> (message from TMGB on Mon, 23
	Aug 1999 11:10:33 -0400)
References: <m3ogfz9mx6.fsf@intergate.bc.ca> <37C16469.687EDE58@am.appstate.edu>
Message-ID: <26081999.4@sanctum.jae>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

>>>>> "TMGB" == TMGB  <bennettt at am.appstate.edu> writes:

    TMGB> In addition to Mr. Clarke's suggestion, seeing that
    TMGB> cool-looking is in the eye of the beholder I'll just suggest
    TMGB> an implementation you may want to look at.  If you are using
    TMGB> Linux 5.2 (I'm not sure about other versions) you may want

I assume you mean Red Hat 5.2?  You know, there *is* more to life than
just RH...  like Debian, for example <0.5 wink>.

I just had to clarify this (as a very satisfied Debian user since
release 1.2 (2.5 years ago))

Bye, J

PS: JAE's Pet Peeve #327: People losing sight of the fact that there
are other Linux distributions than their own.

PPS: This machine here runs potato (the unstable next release of
Debian).  Frankly, it seems more stable than some of the "stable"
releases of other distributions out there.  ;-)

- -- 
J?rgen A. Erhard      eMail: jae at ilk.de      phone: (GERMANY) 0721 27326
         My WebHome: http://members.tripod.com/~Juergen_Erhard
           "Outside of a dog, a man's best friend is a book;
           inside of a dog, it's very dark." --  Groucho Marx
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v0.9.10 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAjfFEiQACgkQN0B+CS56qs3HdACeJRPGX+jK3exE1kly8zSMr6JV
mAoAn145eZrIZ287jg3e+Wz3+NlAung9
=yExc
-----END PGP SIGNATURE-----



From hoel at GermanLloyd.org  Mon Aug  2 11:23:53 1999
From: hoel at GermanLloyd.org (Berthold Hoellmann)
Date: Mon, 02 Aug 1999 17:23:53 +0200
Subject: Libraries needed, please help!!!
References: <7ns6o5$jo2@news.go.dlr.de><004e01beda98$d14f9390$f29b12c2@secret.pythonware.com><7o3moo$dse@news.go.dlr.de> <14245.40279.79936.219772@weyr.cnri.reston.va.us> <7o4ak3$imi@news.go.dlr.de>
Message-ID: <37A5B809.A29DDFF5@GermanLloyd.org>

Sven Drescher wrote:
> Thanks!
> 
> But there are still errors.
> 
> My programm is very little, so I think, there are no syntax errors in it:
> 
> #include "Python.h"
> int main(int argc, char **argv)
> {
>   Py_Initialize();
>   PyRun_SimpleString("print 'Hallo, embedded world!'");
> return 1;
> }
> 
> Ok. I compiled and linked my program with the command:
> 
> CC -I/path/to/include/ -L/path/to/lib/ -o gui gui.cxx
> errors:
> undefined symbols ... first referenced in 'my_object_file.o'
> 
> The undefined symbols Py_Initialize and PyRun_SimpleString are still the
> error messages.
> What did I wrong?
> Can anybody send me the compiler-options for embedding needed object-files
> or other flags? Has anyone written a makefile, to do these work?

Order matters? Try:

CC -I/path/to/include/ -L/path/to/lib/ -o gui gui.cxx -lpython1.5

Good luck

Berthold
-- 
email: hoel at GermanLloyd.org
   )
  (
C[_]  These opinions might be mine, but never those of my employer.



From sholden at pgrad.cs.usyd.edu.au  Sat Aug 14 23:09:30 1999
From: sholden at pgrad.cs.usyd.edu.au (Sam Holden)
Date: 15 Aug 1999 03:09:30 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>  <37b4ce93@cs.colorado.edu> <slrn7r9la9.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7r9nde.h3c.jstevens@bamboo.verinet.com> <slrn7r9p2h.pu8.sholden@pgrad.cs.usyd.edu.au>  <slrn7rb0uh.cf9.jstevens@bamboo.verinet.com>
Message-ID: <slrn7rcbs7.rsf.sholden@pgrad.cs.usyd.edu.au>

On Sat, 14 Aug 1999 14:57:21 GMT,
    John Stevens <jstevens at bamboo.verinet.com> wrote:
>On 14 Aug 1999 03:36:23 GMT, Sam Holden <sholden at pgrad.cs.usyd.edu.au> wrote:

>Your example included:
>
>@array = (1,10,20,30);
>$from_list = (1,10,20,30);
>$from_array = @array;
>print "$from_list\n$from_array\n";
>
>The line you specify does not contain a list, it contains a
>tuple.
>
>>No you can't.  Is there an array in that line of code?
>
>No, and that isn't a list, even if you call it one.  It is a tuple.
>
>>No. Is there a list
>>in that line of code?
>
>No, it's a tuple, not a list.

You can play semantic games all you like, here's the only definition of
a tuple that I could find :

tuple :
In functional languages, a data object containing two or more
components. Also known as a product type or pair, triple, quad, etc.
Tuples of different sizes have different types, in contrast to lists
where the type is independent of the length. The components of a tuple
may be of different types whereas all elements of a list have the same
type. Examples of tuples in Haskell notation are (1,2), ("Tuple",True),
(w,(x,y),z). The degenerate tuple with zero components, written (), is
known as the unit type since it has only one possible value which is
also written ().

Sorry perl's lists are not tuples. Since they can only store values of
one type - scalars. You can't have a list of lists (sorry tuple of tuples),
only a list of scalar refernces to a lists (sorry tuple of scalar references
to tuples).

Perl uses the word list. The word list has many more meanings in the English
language than the Mathematics definition I'm sorry. Perl uses a different
meaning than you like, go and take it up with whoever publishes your
dictionary.

>Go ahead, show me a variable that contains a list.  In your
>example:

Did you read my post. Particularly the bit in which I said you can't have
a list in a a variable. Anyway I give up - learn how to read, and accept
that perl uses the work list differently than python.

>
>$from_list = (1,10,20,30);
>
>$from_list isn't a list.
>
>>Again I repeat, here is a perl list : ('a', 'b', 'c') or qw(a b c)
>
>Again, you are wrong.  The construct ('a', 'b', 'c') is a tuple.
>Perl'ers may call it a list, but how do you perform a an insert
>on the above?

Strangely enough perl didn't involve in an ivory tower. It isn't meant to
be a 'pure' language, it doesn't have lots of 'theory' behind it. Here's a good
definition of a list that the paper boy will tell you is a list :

list n 1: a database containing an ordered array of items (names or
topics) [syn: listing] 2: the property possessed by a line or surface
that departs from the vertical; "the tower had a pronounced tilt";
"the ship developed a list to starboard"; "he walked with a heavy
inclination to the right" [syn: tilt, inclination, lean, leaning] v
1: give or make a list of; name individually; give the names of 2:
include in a list 3: give the names of; "Name the states west of the
Mississippi!" [syn: name] 4: enumerate; "We must number the names of the
great mathematicians" [syn: number]

no mention of operations there. When I ask for a list of birthdays, I don't
want something I can insert into, or delete from. I want a bunch of dates
written on a peice of paper. Feel free to ask for a tuple of birthdays, I'll
stick you common English though thanks.

>
>You don't.  You copy the contents of your tuple into an array,
>then perform array operations on that array.
>
>>>I know Perl.  You need to learn Python.
>>
>>Did I mention a single bit of python syntax in my post? No. 
>
>So what?  I can still make a comment.  And if you had been
>paying attention, you would have realized the context of this
>thread.

And if you read my post you would have realised I avoided comparing perl
and python in any way.

>
>>What was the point of saying I need to learn Python?
>
>Because, in the words of TC, "Learning is a good thing".
>
>More specifically, had you studied Python, it would have been
>clear that what Perl calls a list is more properly refered to
>as a tuple.

I agree with you. An earlier post by TomC specifically said that. Perl
calls them lists, thus they are lists. If Perl called them Hamburgers then
they would be hamburgers. Of course that would make perl harder to talk
about and to understand from the documentation.

Show a bunch of people the following (try paper boys, programmers,
mathematicians, etc) :

1,2,3,4,5,6,7,8,9

Ask them what it is.

I bet more people say a 'list' than say a tuple.

Perl didn't come from an ivory tower, it uses English words for their 
English meanings a lot. You don't like that, fine, just don't argue perl
doesn't have lists when you know it does. 

If you had said "perl doesn't have lists, what it calls lists are really
tuples', I would have answered with aomething like the above. I misunderstood
and thought that you thought arrays and lists where exactly equivalent in perl.
That is a common mistake in perl, that people who make the mistakes that you
have made with perl code in this thread make. People who struggle with
using @ where they mean $ and {} where they mean (), etc tend to be
misunderstanding the basic data structures perl provides.

I made a mistake, I can admit that.

I do have a feeling that you knew I was and kept going anyway because you
like to argue about stupid pointless things.

>
>>You might as well have told me to learn C or Lisp.
>
>Yes.  You should.  Had you learned Lisp, list processing would
>be second nature to you, and you would have a clearer idea between
>"lists" and "tuples" and "arrays".

In fact I have. But the meaning of 'list' when talking about perl is whatever 
perl calls a list. Ignore it all you want. You must get misunderstood a lot.

>
>Stop.  You are talking about arrays, not lists.  I repeat, if
>@ary were a list, the above operation would be illegal.

But @ denotes an array and you know this. 

>
>>Of course you know better then perl itself.
>
>In this case, I would say that Perl should use the same terminology
>as other fields of study.  So, yes, I would consider that Perl
>makes a mistake, calling this a list.

That's fine, your allowed to disagree with the terminology. but since you
in all likelyhood knew what I meant anyway you have just been playing games.


*plonk*

-- 
Sam

I explicitly give people the freedom not to use Perl, just as God gives
people the freedom to go to the devil if they so choose.
	--Larry Wall



From gdupuis at rocler.qc.ca  Thu Aug 19 11:19:37 1999
From: gdupuis at rocler.qc.ca (Gabriel Dupuis)
Date: Thu, 19 Aug 1999 15:19:37 GMT
Subject: The stupidest question ever seen in this newsgroup...
References: <37BB8F9A.6C204793@rocler.qc.ca>
Message-ID: <37BC1ECB.C2D3C602@rocler.qc.ca>

Gabriel Dupuis wrote:
> 
> Yeah, I know that it's a shame to ask a question like this one but...
> 
> Ok, I just bought "Programming Python" in the hope to learn a superb
> programming langage like Python. But the first program that I try from
> the book doesn't work!
> 
> Here's the source : (as you can see I haven't been into programming for
> a while)
> 
> #!/usr/bin/python
> 
> x = y / 2
> while x > 1:
>     if y % x == 0:
>          print y, 'has factor', x
>          break
>     x = x - 1
> else:
>      print y, 'is prime'
> 
> But when I try to run it from my GNU/Linux box I got this error :
> 
> Traceback (innermost last):
>   File "./factor.py", line 3, in ?
>     x = y / 2
> NameError: y
> 
> Pretty sad huh? I know it's probally something very dumb (don't flame me
> please) but I can't simply find the bug and I copied the source from the
> book without any alterations...
> 
> If somebody could just take 30 secs and look at this I'll appreciate
> much more than you can imagine.
> 
> Thanks... (and remember kids, there's no stupid question, just stupid
> people :< )

Thanks alot everybody



From fredrik at pythonware.com  Mon Aug 30 03:46:49 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 30 Aug 1999 07:46:49 GMT
Subject: Question: lists in classes
References: <37CA1803.2FA9261B@ug.cs.su.oz.au>
Message-ID: <011301bef2bb$d3f29850$f29b12c2@secret.pythonware.com>

Jeff Turner <jturner at ug.cs.su.oz.au> wrote:
> Apologies if this has been answered before...

it has...
 
> class foo:
>     list=[]
> 
> a=foo()
> b=foo()
> a.list.append(123)
> print b.list
> 
> I would expect the above code to return "[]", since b has not been
> modified since instantiation.

but it has.  a.list is a class attribute, and are
modified in place.

> However it returns [123].

> How would I tell python that I only want to _define_ the class, not
> instantiate it?

use an __init__ method, just like the rest of us ;-)

> class foo:
>     def __init__(self):
>         self.list=[]

also see:
http://www.python.org/doc/FAQ.html#6.24
http://www.python.org/doc/FAQ.html#6.25

and perhaps
http://www.python.org/doc/FAQ.html#4.84

</F>




From mstenber at cc.Helsinki.FI  Thu Aug  5 22:21:53 1999
From: mstenber at cc.Helsinki.FI (Markus Stenberg)
Date: 06 Aug 1999 05:21:53 +0300
Subject: Stupid Question?
References: <Bioq3.214$6E5.3247@typ12.nn.bcandid.com>
Message-ID: <al8so5xk5ry.fsf@sirppi.helsinki.fi>

"jim moore" <indiana at ev1.net> writes:
> Is there a bulit in function that will give you N evenly spaced numbers
> between X1 and X2.  For example: 10 numbers between 0 and 1 ([0.1, 0.2, 0.3,
> ...])

Uh, sounds pretty trivial to me: 

def linspace(v1,v2,n):
	return map(lambda x,v1=v1,v2=v2,n=n:\
		   v1+float(v2-v1)*(1+x)/(n+1), range(n))

I don't see why everything should be builtin :P

> It would be similar to MATLAB's linspace.
> How about a function similar to range but allowing a floating point step?
> Seems like there would be one out there, so I thought I'd check before I
> wrote one.

Same thing; range's quick-and-dirty integer thingy, if there was float
checks or stuff for the step, it'd slow down slightly or great deal, and I
don't see much of a gain in it. 

> Thanks,
> Jim Moore

-Markus

-- 
"Chicks will dig you if you use Debian!  Running RedHat is
like going to the beach to scope out the action while wearing
your water wings."
        -- slashdot comment



From pj at sgi.com  Fri Aug 27 21:23:24 1999
From: pj at sgi.com (Paul Jackson)
Date: 28 Aug 1999 01:23:24 GMT
Subject: Why do we call python a scripting language?
References: <19990827112918.12652.rocketmail@web129.yahoomail.com> <8E2F958DCPaCmAnRDLM@news.dial.oleane.com> <37c726c4.2148316@news.netset.com>
Message-ID: <7q7dmc$5t01g@fido.engr.sgi.com>

|> Is any one using freeze or ...

SGI will soon post to www.sgi.com a couple more Y2K scanning
tools, one that checks what you have installed on Irix systems
for recommended y2k compliant versions, and one that will
'grep' through source code for y2k sensitive terms.

The 2nd of these, the source scan, I wrote in Python using freeze.

It was the first production Python I have written.  'Twas a
little silly, turning a few hundred line 'find ... | xargs
grep ...'  shell script into a 700 Kbyte executable.  But other
than that, worked fine.

-- 

=======================================================================
I won't rest till it's the best ...	   Software Production Engineer
Paul Jackson (pj at sgi.com; pj at usa.net) 3x1373 http://sam.engr.sgi.com/pj



From tchrist at mox.perl.com  Thu Aug 19 16:24:35 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 13:24:35 -0700
Subject: More random python observations from a perl programmer 
References: <14268.15201.81468.481452@bitdiddle.cnri.reston.va.us> <199908191713.LAA02152@jhereg.perl.com> <14268.22374.331241.721504@anthem.cnri.reston.va.us>
Message-ID: <37bc59f3@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, bwarsaw at cnri.reston.va.us (Barry A. Warsaw) writes:
:>>>>> "TC" == Tom Christiansen <tchrist at jhereg.perl.com> writes:
:    TC> I only have the Lutz book.  I don't have manpages. :-(
:
:That's okay, I only have the first edition Programming Perl book :)

No biggie:  It's nearly all there as manpages. :-)

The Perl Cookbook, on the other hand, is not, although you can ftp all
its code examples and programs, just not its prose.  However, I do have
a policy of giving away Perl books to visitors who show up in Boulder.
I'm sure I could fix that then. :-) (I don't bring a stash to the Open
Source Conference, or I would then.)

--tom
-- 
Those who do not fight the demons within themselves are destined to fight them 
 outside themselves...



From abigail at delanet.com  Mon Aug 23 10:05:32 1999
From: abigail at delanet.com (Abigail)
Date: 23 Aug 1999 09:05:32 -0500
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu> <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net>
Message-ID: <slrn7s2lb3.k6b.abigail@alexandra.delanet.com>

xah (xah at weborder.com) wrote on MMCLXXXIII September MCMXCIII in
<URL:news:37c126a6$0$813 at newsreader.alink.net>:
"" 
"" However, the ivy tower class of buttheads insist on symbolism for math,
"" clinging to the fancy that certain ideas are better expressed in certain
"" language. They created a whole barrier of entry. You couldn't believe what
"" kind of perverted use of symbols they invent. I know because I've been there
"" a bit. Their one-language ideology taints our programing world. If you
"" wonder why so many languages uses weird syntax (e.g. python, dylan, lisp,
"" haskell, mercury, mathematica ...ad nausea), you have these guys to thank
"" to.

Yeah, those math texts from the 16th and 17th century, they are sooooo easy
to read, because they lack all the symbols. Not to mention the original
Euclid and those other Greeks, who had no symbols at all!

"" On a personal note...  Do you know what a computer science degree is these
"" days? They require you understand shits like lambda calculus, logic, graph
"" theory, combinatorics, or even algorithms?? All for what? All the Perl
"" programers I know doesn't understand any of these, and they all are making
"" big bucks in fortune 500. I'm not ashamed that I dropped out of school. I
"" hope that Larry and Tom's teachings will eventually make such subjects
"" disappear for good.


Hahahahahahahahahahhahahahahaahahahahahaha.


If you think there's much relationship between programming and Computer
Science, you're utterly mistaken. It's the same relationship between
driving your Morris Minor and designing and producing a new car.

Larry and Tom can teach all they want; their teachings will have no impact
on Computer Science what so ever. Just like their teaching will have no
impact on the grow of potatoes in Idaho.



Abigail
-- 
perl -we 'print split /(?=(.*))/s => "Just another Perl Hacker\n";'


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From S.I.Reynolds at cs.bham.ac.uk  Thu Aug 26 13:47:23 1999
From: S.I.Reynolds at cs.bham.ac.uk (Stuart Reynolds)
Date: Thu, 26 Aug 1999 18:47:23 +0100
Subject: count method only for lists?
Message-ID: <37C57DAB.63C7@cs.bham.ac.uk>

Does anyone know why there the count method is only defined for mutable
sequences (i.e. lists) and not strings and tuples? And for that matter,
index too? It seems as though the Python Library Reference (2.1.5.2
Mutable Sequence Types) only describes lists and not mutable sequence
types.


Stuart



From ralf.doering at prakinf.tu-ilmenau.de  Mon Aug  9 06:43:40 1999
From: ralf.doering at prakinf.tu-ilmenau.de (Ralf Doering)
Date: 09 Aug 1999 12:43:40 +0200
Subject: rexec question
References: <xz4sievhzv.fsf@prakinf.tu-ilmenau.de> <14249.56828.918361.13395@bitdiddle.cnri.reston.va.us> <xzg11xwf10.fsf@prakinf.tu-ilmenau.de> <14251.13832.844951.142519@bitdiddle.cnri.reston.va.us>
Message-ID: <xz4si99qub.fsf@prakinf.tu-ilmenau.de>

Jeremy Hylton <jeremy at cnri.reston.va.us> writes:


[...]

THX again for the response. It does help me to understand what
happens, but I need a little bit more help to solve my problem.

> I should have answered this question from the outset.  The restricted
> execution environment is definitely more interesting when there are
> objects defined in the controlling environment that are all accessible 
> in the restricted environment.
> 
> There is no direct way to add a name binding to the restricted
> execution environment.  You need to modify one of the modules that
> exists in rexec, typically its __main__.
> 
> Here's a simple approach that would work:
> 
> class RExecEx(rexec.RExec):
>     """Adds method for setting attribute in rexec environment"""
>     def setattr(self, name, value, bastion=1):
>         if bastion:
>             value = Bastion.Bastion(value)
>         setattr(self.modules['__main__'], name, value)
> 
> In most cases, you want to pass a Bastion to the untrusted code rather 
> than passing the object directly.  The howto and the Bastion module
> documentation explain this.

OK, I made an "setattr" enhanced version of rexec like suggested.

To illustrate what I do not understand until now:

>>> class t:
...	def t1(self):
...		print "open is:", open
... 
>>> r = RExecEx()
>>> r.r_exec("""print "open is", open""")
open is <method RExec.r_open of RExecEx instance at 1fa7e8>
>>> o = t()
>>> r.setattr("o",o)
>>> r.r_exec("o.t1()")
open is: <built-in function open>


The problem to import from the controlling environment to the
restricted environment is solved. However, the restricted execution of
the code of the "imported" object does not work like expected.
The call to o.t1() uses the builtin open function instead of r_open.
Please, can somebody enlighten me ...
Again, the problem is: I have got an object, which is unpickled from a
bytestream. A specific method of this object should be called and the
code should be executed in an restricted environment.

Ralf
-- 
No sig -- no fun



From arcege at shore.net  Mon Aug 30 16:57:11 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 30 Aug 1999 20:57:11 GMT
Subject: A Plea For A Return To Sanity!!!
References: <Pine.LNX.4.10.9908281230060.9521-100000@starship.python.net> <7qehg2$bkc@dfw-ixnews4.ix.netcom.com>
Message-ID: <HeCy3.662$1k5.226217@news.shore.net>

Aahz Maruch <aahz at netcom.com> wrote:
: In article <Pine.LNX.4.10.9908281230060.9521-100000 at starship.python.net>,
: Moshe Zadka  <moshez at server.python.net> wrote:
:>On Sat, 28 Aug 1999, Tim Peters wrote:
:>>
:>> python-folk-are-the-best-buddies-psu-dues-can-buy-ly y'rs  - tim
:>                                   ^^^
:>There are a few options:
:>
:>1. The mailman at Python org had a glitch
:>2. The timbot made a type
:>3. There's this underground association, Python Software Union, which goes
:>around breaking thumbs of people who use Python without paying the PSA.

: Sheesh, does nobody here remember psuvm?

I remember it..  remember pizzafest? :)

  -Arcege




From dave at see.sig.for.address  Sun Aug  8 16:46:56 1999
From: dave at see.sig.for.address (Dave)
Date: Sun, 08 Aug 1999 20:46:56 GMT
Subject: downloading web based email
References: <7och4d$jr0$1@nnrp1.deja.com>
Message-ID: <37ade9b3.3531069@read.news.global.net.uk>

On Thu, 05 Aug 1999 17:20:26 GMT, Preston Landers <mithy at my-deja.com>
wrote:

> Hello everyone....
>
>I'm thinking of writing a Python program to download email from
>web-based services such as Yahoo mail, Deja, Hotmail, and so on, and
>deliever the mail in a format that standard unix mailers can use.
>
>I'm posting here to make sure that this work has not already been done.
>If someone has already seen a program like this, that can be adapted for
>a variety of web based email services, please let me know.  I'm not
>writing this as a learning exercise,  I just need the tool. ;-)

Its not in python, but if you are on a Windows system there is a
program called CWebMail (at http://www.cwebmail.com) that does pretty
much as you ask for Yahoo and hotmail. It acts as a proxy server so
you can download the mail into your normal mail program.


	Dave K

--------------------------------------------------
All great ideas start as heresy and end as dogma.

dkirby@   <-figure this out, spambots!
bigfoot.   My opinions are my own,
com        but I'm willing to share.



From fw at s.netic.de  Fri Aug 20 07:00:37 1999
From: fw at s.netic.de (Florian Weimer)
Date: 20 Aug 1999 13:00:37 +0200
Subject: More random python observations from a perl programmer
References: <37BC6986.A66B044E@starmedia.net> <199908191634.KAA01333@jhereg.perl.com>
Message-ID: <87d7wiaf8q.fsf@deneb.cygnus.qad.org>

Tom Christiansen <tchrist at jhereg.perl.com> writes:

> >Try reading the docs...
> 
> Thanks for nothing: there are no manpages.

Manpages are most useful for languages which do not support namespaces
(so you can hit `K' or `M-x man RET RET', and you get the right page).
Perl fits into this more than Python (which heavily uses methods on
builtin objects).  Unfortunately, `man exists' (or `perldoc exists')
doesn't work with my Perl installation either, so I don't see why it
makes that much sense to have any manpages.

In addition, it's a PITA to search for a specific function in the
`perlfunc' manpage because the entries aren't marked at all.  You have
to prepend a few spaces to the function name and search for that,
otherwise you might get too many references in the descriptions of other
functions.  I must admit that personally, I like Python's structured
Info documentation much better.



From init__self at my-deja.com  Mon Aug 30 16:01:10 1999
From: init__self at my-deja.com (Python Addict)
Date: Mon, 30 Aug 1999 20:01:10 GMT
Subject: All rise for judge Timothy Peters...
Message-ID: <slrn7sloo9.r4u.init__self@jules.com>

I kid you not. I'm involved in a legal battle (on the prosecution
side, thank you very much) and spent a morning in court just
recently. My flabbers where ghasted when I noticed the Judge's name
plate.

Now "our" Tim always goes at length to not explain what he does so
I'm wondering if these two are the same person...

-- 
Why join the navy when you can be a pirate?




From volker.hetzer at abg1.siemens.de  Wed Aug 25 09:51:30 1999
From: volker.hetzer at abg1.siemens.de (Volker Hetzer)
Date: Wed, 25 Aug 1999 15:51:30 +0200
Subject: Python 1.5.2 on HPUX 10.20
References: <37C3D5BD.72024971@abg1.siemens.de> <19990825140816.R28043@xs4all.nl>
Message-ID: <37C3F4E2.4953BD78@abg1.siemens.de>

Thomas Wouters wrote:

> > Is this serious? What am I to do?
[info that the test has to do with locking.]

So I can safely ignore it.
Thanks a lot!
Volker
-- 
Hi! I'm a signature virus! Copy me into your signature file to help me spread!



From tchrist at mox.perl.com  Fri Aug 20 09:04:58 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 20 Aug 1999 06:04:58 -0700
Subject: More random python observations from a perl programmer 
References: <37BC6986.A66B044E@starmedia.net> <199908191634.KAA01333@jhereg.perl.com> <012d01beeae7$f3646320$f29b12c2@secret.pythonware.com>
Message-ID: <37bd446a@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    "Fredrik Lundh" <fredrik at pythonware.com> writes:
:as so often before, Tom ran out of positive energy and
:started posting stuff like:
:> You seem to be looking for a fight.   Please say nothing
:> at all if that's all you want to do.

Thank you for your constructive encouragement and enthusiasm.  

Like most people, I do not take kindly to digs.  Would that
I were so patient as thou.

--tom
-- 
"A young man who is not a radical about something is a pretty poor risk
 for education."
				- Jacques Barzun



From fthiriet at coming.fr  Mon Aug  9 07:29:42 1999
From: fthiriet at coming.fr (Frederic Thiriet)
Date: Mon, 9 Aug 1999 12:29:42 +0100
Subject: Problems with Grail 0.6 and Python 1.5.2 (NT)
References: <933940498.641759@mhs0096x.mchh.siemens.de>
Message-ID: <7omaa2$36e$1@jaydee.iway.fr>

I've got exactly the same configuration NT 4 sp5, python 1.5.2, grail 0-6
and it worked well . I had to modify the registry to call the *.py files
directly from the DOS shell, as you do.

Have you tried

> python grail.py

just an idea....

BTW, I find that grail 0.6 is still not really useable, although it is a
very intersting project.





From hyoon at bigfoot.com  Thu Aug  5 16:26:14 1999
From: hyoon at bigfoot.com (Hoon Yoon)
Date: Thu, 05 Aug 1999 16:26:14 -0400
Subject: ADO example code failed:hlp!
Message-ID: <37A9F366.EA01D915@bigfoot.com>

Running Christian's code, I get following error message. What am I doing
wrong? It's same code, but the constants do not contain adModeReadWrite.
Pls help!
Thanks much in Adv,

import win32com.client
db=win32com.client.Dispatch("ADODB.Connection")
from win32com.client import constants
db.Mode = constants.adModeReadWrite
db.Open("Access DB")
db.Execute("create table mist (name varchar, wert long)")
rs=win32com.client.Dispatch("ADODB.Recordset")
rs.Open("mist", db, constants.adOpenDynamic, -1, constants.adCmdTable )
rs.AddNew()
rs.Fields("name").Value="hugo"
rs.Fields("wert").Value=123.45
rs.Update()

Python 1.5.2 (#0, Apr 13 1999, 10:51:12) [MSC 32 bit (Intel)] on win32
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> ## working on region in file c:\tmp\python-aDA-KJ...
Traceback (innermost last):
  File "<stdin>", line 1, in ?
  File "c:\tmp\python-aDA-KJ", line 4, in ?
    db.Mode = constants.adModeReadWrite
  File "C:\Python\win32com\client\__init__.py", line 72, in __getattr__
    raise AttributeError, a
AttributeError: adModeReadWrite
>>> 
-- 
*****************************************************************************
S. Hoon Yoon (Quant) Merrill Lynch Equity Trading, 
yelled at yahoo.com hoon at bigfoot.com(w)
"Miracle is always only few standard deviations away, but so is
catastrophe."
* Expressed opinions are often my own, but NOT my employer's.
"I feel like a fugitive from the law of averages."    Mauldin
*****************************************************************************
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vcard.vcf
Type: text/x-vcard
Size: 202 bytes
Desc: Card for Hoon Yoon
URL: <http://mail.python.org/pipermail/python-list/attachments/19990805/c79a9fba/attachment.vcf>

From shecter at darmstadt.gmd.de  Thu Aug 19 12:36:10 1999
From: shecter at darmstadt.gmd.de (Robb Shecter)
Date: Thu, 19 Aug 1999 18:36:10 +0200
Subject: Kudos to documentors: (popen2)
References: <37BBF0C2.C0E8B3C8@darmstadt.gmd.de>
		<37bc15c2@cs.colorado.edu>
		<37BC21D9.E19DBEA2@darmstadt.gmd.de> <14268.10767.95070.208607@weyr.cnri.reston.va.us>
Message-ID: <37BC327A.3DA151EA@darmstadt.gmd.de>

"Fred L. Drake, Jr." wrote:
>... I'd like to point out that
> ...it is based on
> language semantics and not on presentation.  

Hi,

That's what I was trying to say!  I think that my writing was not too
clear - I was trying to say that the Java and Python doc systems
should be used as examples for other languages.

It'd be cool if parameters and return values were hyperlinked to the
definitions of their types, but I don't think that'd be possible in a
dynamically typed language.  Maybe this isn't such a big problem for
programmers who are used to working with it.

It did bug me, though, because my scenario looked like this:

1) I wanted to start a subprocess and get its stdout and stdin.  Using
the library reference and indexes, it was very easy to find popen2.

2) From that page I got the module and method name I needed.  Great.

3) I needed to know how to work with the return values.  The page says
"Returns (child_stdout, child_stdin)".  I got stuck here.  Without the
type, I didnt know how to read/write to them.  What I eventually did
was start a Python interpreter, do a method call, and print the
returned objects - which told me their types.  After -that-, I could
go back to the docs and see how to read and write to them.

It would have been great if their return types were listed, along with
hyperlinks to the appropriate docs.  But that wouldn't come for free
in a dynamically typed language, I guess.

Maybe my strategy for figuring the class out was incorrect... I do
most of my work in statically typed languages, so that may have an
influence.

- Robb



From arcege at shore.net  Mon Aug  9 14:37:15 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Mon, 09 Aug 1999 18:37:15 GMT
Subject: Python, Embedding Q's
References: <7odrn8$qf2$1@wyrm.its.uow.edu.au>
Message-ID: <vdFr3.1216$EG4.199798@news.shore.net>

B Kingsford <bryn at mds01.itc.com.au> wrote:
: hi all

: im embedding Python in multi-threaded server, and just had a
: couple of q's regarding compiling files, cache, etc. if any help
: or thoughts could be offered, please do.

: I wish to use Python's built-in threading of the interpreter,
: aswell as Python's memory management [for portability], will
: this have any major, adverse performance effects?
: Does Python handle which thread does what?
: would it be better to create say, x threads, each with their
: own instance of the interpreter?

If you embedded Python in the threads, then yes, each thread could/
would have its own interpreter, but Python can handle multiple threads
itself.  Python will handle the memory management for you; access to
Python is serialized, but non-Python code is free to be threaded.

Read up on the threading API:
  http://www.python.org/doc/current/api/threads.html

: what is the PyDict format of globals/locals from PyModule_GetDict?
: what vars are declared in this ModuleDict? is it necessary to add
: anything there?

The basic variables would be "__name__", "__builtins__" (bound to the
__builtin__ module), "__doc__" and possibly the "__file__" attribute.
Other variables you want to add at initialization should be by calling
  PyDict_SetItemString(moddict, "varname", object);

You probably only want to access PyModule_GetDict() (which returns a
normal PyDictObject) only when you initialize the module, and use
PyObject_*AttrString() at other times.

: does PyCodeObject keep (or represent) the compiled code in mem?
: would it be worth maintaining a heap of PyCodeObjects (one for
: each .pyx, or until cache is full...) as opposed to recompiling
: for each request (unless it is desired, say, .pyd), say
: PyCodeObject  * pcoCached[] = malloc( 500000 );
: will Py use this new heap for the CodeObject pointers aswell as?
: will the CodeObjects fill the heap, or will Py, when compiling
: for a CodeObject, create mem elsewhere? eer, does sizeof(PyCodeObject)
: include the Py bytecode?

The code objects will be wrapped around functions and bound to
variables in the modules you import.  You can then call the functions
with PyObject_CallFunction() or from PyRun_SimpleString(), depending on
how you are using the API.

It is probably better to let Python maintain the information for you,
and it will take care of the memory.

: if i were to maintain the PyCodeObjects, how then would i have
: an instance of the interpreter run this code? ahh, obviously
: PyEval_EvalCode, but what if Python is handling the sub-interpreters?
: how to say, ThatSubInterpreter->EvalCode( PyCodeObject )?

This is handled by switching the interpreter state (see the webpage
mentioned above).  But you will want to run the code as functions.  For
a good example of switching interpreter states, look at the _tkinter.c
code.

: is Py_NewInterpreter really a 'fresh copy' of the main interpreter
: for use by a thread? if i add built-in mods with PyImport_AddModule,
: do i need to do this for each Py_NewInterpreter? what about
: Py_InitModule? does this need to be called for each new aswell?

As far as I understand it looking at the code, yes, but I've never used
Py_NewInterpreter(), so maybe someone else can answer better.  Each new
interpreter has it's own set of imported modules, so yes, you will need
to import the modules individually.  But the Py_InitModule should be
called from the modules "init<modname>" function (through a normal
"import <modname>" statement).
  http://www.python.org/doc/current/ext/mehodTable.html

: in the examples, the telnet server seems to call Py_NewInterpreter
: for each request. would i have to, if say i wished to limit the
: worker threads, when a request occurs, if threads >= max, put
: request in queue, OR, create a new thread with Py_NewInterpreter,
: handle the request, check to see if any new requests are in the
: queue [and handle them too], then Py_EndInterpreter? what about
: reusing Py_NewInterpreters after they've been idle for a while?
: (say, a LIFO q of interpreters... you know how it is... no action
: for a while then bang, everybody wants in on your server!!)

If this is the case, it would probably be better to create a thread
pool and work from a queue.  There are other techniques you could use
tho.

  -Arcege




From moritz at news.uni-mannheim.de  Fri Aug 20 13:30:12 1999
From: moritz at news.uni-mannheim.de (Moritz Moeller-Herrmann)
Date: Fri, 20 Aug 1999 19:30:12 +0200
Subject: More random python observations from a perl programmer
References: <37BC6986.A66B044E@starmedia.net> <199908191634.KAA01333@jhereg.perl.com> <87d7wiaf8q.fsf@deneb.cygnus.qad.org> <37bd44ed@cs.colorado.edu>
Message-ID: <slrn7rr454.1h0.moritz@Gertrud.chaos.wg>

On 20 Aug 1999 06:07:09 -0700, Tom Christiansen <tchrist at mox.perl.com> wrote:
>     [courtesy cc of this posting mailed to cited author]

>In comp.lang.python, 
>    Florian Weimer <fw at s.netic.de> writes:
>:Manpages are most useful for languages which do not support namespaces
>:(so you can hit `K' or `M-x man RET RET', and you get the right page).
>:Perl fits into this more than Python (which heavily uses methods on
>:builtin objects).  Unfortunately, `man exists' (or `perldoc exists')
>:doesn't work with my Perl installation either, 

>I really hate that.  You're right about perlfunc being insanely long.
>I've always argued against it, and there's a script in the distribution
>kid that *will* create things like /usr/local/perl/man/man3/exists.3,
>but I haven't managed to get my lobbying power worked up enough to push
>through to make that the default.

Maybe you will like a better info file viewer? Have a look at  Przemek Borys'
<pborys at dione.ids.pl> pinfo, a much better replacement for the official gnu
info viewer.  It will also read manpages and display them. Freshmeat has the
URL.

And I hate Emacs :-)


Just noted that you ought to love this guy's email address...


-- 
Moritz Moeller-Herrmann mmh at gmx.net ICQ# 3585990        # Not only
Get my public pgp / gpg key from                        # Open Source(TM)
http://webrum.uni-mannheim.de/jura/moritz/pubkeymoritz  # but also
KDE forever! Use Linux to impress your friends!         # Open Minded!




From jstevens at basho.fc.hp.com  Fri Aug 13 22:53:48 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 19:53:48 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132029.OAA03987@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:29 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cc2c@cs.colorado.edu>

> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :Perl actually seems to be a *SURRENDER* to real-live-languages.  This
> :makes [YOUR NEWSREADER SCREWED THIS UP]
> :it easier to code in *YOUR* way
> :(there-is-always-more-than-one-way-to-do-it), [AND THIS ONE]
> :but this (just like natural language) provides greater opportunities for
> :confusion and ambiguity in the mind of the reader.
> 
> Perhaps then the problem is actually the plummeting literacy rate.

Oh, I doubt it.  The biggest problem in management is the requirement
to use a natural language to manage people (allowing for far to many
ambiguities and mis-understandings).

Illiteracy doesn't help, of course.  But I don't think it is the
root problem.

John S.




From bernhard at alpha1.csd.uwm.edu  Tue Aug 31 12:43:33 1999
From: bernhard at alpha1.csd.uwm.edu (Bernhard Reiter)
Date: 31 Aug 1999 16:43:33 GMT
Subject: alien editor in PythonWin/IDLE
Message-ID: <slrn7so1hl.s1c.bernhard@alpha1.csd.uwm.edu>

Just wondering,

how difficult would it be to integrate your favourite editor into
IDLE and PythonWin?

I haven't looked at it yet, but should'nt there be a uniform interface
to integrate other editors.

For example I'm using gvim on Windows to write python.
But emacs and nedit might be interesting for IDLE, too.

Just to raise the thought.
	Bernhard
-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.    (www.uwm.edu/~bernhard)
Association for a Free Informational Infrastructure              (ffii.org)



From andrew at starmedia.net  Thu Aug 19 17:24:00 1999
From: andrew at starmedia.net (Andrew Csillag)
Date: Thu, 19 Aug 1999 21:24:00 GMT
Subject: More random python observations from a perl programmer
References: <199908191634.KAA01333@jhereg.perl.com>
Message-ID: <37BC75F0.B482DA6D@starmedia.net>

Tom Christiansen wrote:
> 
> >> How is that a singleton tuple?
> >It's not.  For format strings, if the item (here 4.5/100) is a single
> >thing, the items need not be a tuple. i.e.
> 
> That's painfully inconsistent.

No, just extremely convienent.  Why have to have a single item tuple for
a single item?  It's not that inconsistant, there are two options for
it:  either a single item, or a tuple of items (which may be one item). 
The only time I get tripped up by this is when I do things like:
    foo = "some format string %s" % b
and b turns out to be a tuple, but for me, at least, it happens
infrequently and is easily fixed by using a 1 item tuple instead:
    foo = "some format string %s" % (b,)
or since I'm using %s anyway,
    foo = "some format string %s" % str(b) 

> 
> >>     print "You have $%0.2f bucks" % (4.5/100,)
> >> illegal,
> >No, it's perfectly fine.  Didn't try it, I guess.
Not an attack, just a clarification.

> 
> No, I thought I had.
> 
> >>     print "You have $%0.2f bucks" % ((4.5/100,) * 2)
> >This, on the other hand is wrong since ((4.5/100,) * 2) yields (0.045,
> >0.045) and you only have one formatting bit in the string (only 1
> >%0.2f).
> 
> Typo.
> 
> >> mandatory?  Was this (elt,) thing just an, er, unforeseen design
> >> misfeature? Yes, I tripped me up again. :-(
> >Try reading the docs...
> 
> Thanks for nothing: there are no manpages. 
Manpages don't work very well if you're stuck on a Windows box :(.  If
only the world were Unix... <sigh>

There are the python manuals which are excellent, available at:
http://www.python.org/doc/ (linked to from the homepage)
available in your choice of HTML, PostSript, PDF, Info, LaTeX and
windows help.

Not to mention that they're searchable via:
http://www.python.org/search/

But for the most part, I usually don't need the search portal since the
indexes are really good.  But if you want a greppable form, there's Info
(although I don't like info in general, but info (both standalone and
emacs versions) does have the document search key 's'), HTML, and the
LaTeX versions grep quite well.

> I read the Lutz book.

I personally think that the Lutz book (the big thick one) isn't all that
great (sorry Mark).  Not an offence to you, I bought it also.  The
online tutorial:
http://www.python.org/doc/current/tut/tut.html
last I looked is quite good.  The Learning Python book seems to be quite
a bit better.  The pocket reference is also quite handy.

> You'll note that I even MENTIONED this issue in a gotcha.
> It's just easy to get tripped up by the inconsistency.
> 
> You seem to be looking for a fight.   Please say nothing
> at all if that's all you want to do.

No, actually, just trying to help you out, so chill out alright?  I've
got better things to do then get in a flame war with you over nothing.

Jees, try to be nice and have to go and buy asbestos clothing!

-- 
"Programmers are just machines that convert coffee to executable code"
   -Unknown

"Drew Csillag" <drew_csillag at geocities.com>



From paul at prescod.net  Tue Aug 10 09:30:13 1999
From: paul at prescod.net (Paul Prescod)
Date: Tue, 10 Aug 1999 13:30:13 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com>
		<19990805094020.A1838753@vislab.epa.gov>
		<37aacdf9.324492916@news.oh.verio.com>
		<934117874snz@vision25.demon.co.uk>
		<7om59k$cqc$1@newshost.accu.uu.nl> <934237862snz@vision25.demon.co.uk>
Message-ID: <37B02965.B1FE0E8D@prescod.net>

Have you looked into XUL, XwingML and Glade?

http://www.oasis-open.org/cover/xul.html
http://www.bluestone.com/xml/XwingML/
http://glade.pn.org/

I don't know whether these are exactly the same as what you are doing
but they are certainly related -- and both already have implementations.
Glade's is already in Python!

Phil Hunt wrote:
> 
> It is important to me that the tool is easy to use. This is because
> I intend to use it myself!

Obviously ideas of ease of use vary. Some might say that XML is easier
to use because:

 * all major text editors have a mode for it

 * it can be trivially parsed with existing code written in several
languages

 * APIs for tree manipulation are well known

 * the syntax is (becoming) familiar

 * XML transformation languages (including some nice Python libraries)
can convert your language into XUL and vice versa -- or whatever else
comes along (and I have a feeling several of these interface description
languages will come along ... there will likely be as many as there are
widget toolkits)

You are certainly free to do it however you want. It's your language and
it should be something you are comfortable with!

> Also, IIRC XML is restrictive where it allows comments -- you can't put
> them just anywhere. 

You can put them between elements. That is almost always enough...

> It will also use the C preprocessor, so for example one could
> write a macro for a label and associated textfield:
> 
> #define labelTF(txt,id,size) \
>    label @id##Label txt textField @id cols=size

XML has a similar facility but I think that textual preprocessing is a
little gross whether done in C or XML. I would encourage you to consider
structured alternatives. Whether you use C-syntax or XML syntax, you
could rather have a first-class "object definition" syntax. SVG has such
a thing for defining 2-dimensional shapes.

 Paul Prescod



From skip at mojam.com  Sat Aug 14 17:58:13 1999
From: skip at mojam.com (Skip Montanaro)
Date: Sat, 14 Aug 1999 16:58:13 -0500
Subject: deadlock when two servers call each other - will threads help?
Message-ID: <199908142158.QAA24395@dolphin.mojam.com>

I'm using XML-RPC as the transport mechanism within a group of database
servers to propagate change information and thus maintain database
consistency.  My servers are logically connected in a tree structure.  Each
server has one parent (except the topmost server) and zero or more
children.  When a change request comes in to a server, if it doesn't come
from the server's parent, it forwards it there.  This allows requests to
come in anywhere in the system, then get percolated through all the servers
by forcing it to travel to the top of the tree, then trickle down.

This is fine in theory, but in practice it yields deadlocks.  A child can be
making an XML-RPC call to its parent to forward one change request at the
same time the parent is trying to make another call to that child.  Both
servers will block in read() waiting for a response that never arrives.
System throughput suffers...

I don't use threads at the moment, but I'm curious whether or not threads
would help here.  Would the blocked read() call allow another thread to run?
If so, I could spawn threads to make RPC calls between servers.

Thanks,

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098




From pduffin at mailserver.hursley.ibm.com  Thu Aug 12 04:25:35 1999
From: pduffin at mailserver.hursley.ibm.com (Paul Duffin)
Date: Thu, 12 Aug 1999 09:25:35 +0100
Subject: configure is screwing up threadmodule.c
References: <37B25CD3.E25A4CC5@sprint.ca>
Message-ID: <37B284FF.5690@mailserver.hursley.ibm.com>

Orlando Vazquez wrote:
> 
> hi, i'm having problems getting threads to work in python. my problem is
> at compile time. i have pthreads 1.0 installed on my linux computer
> (slackware 3.4/kernel 2.0.30/gcc v2.7.2.3) when i ./configure python i
> pass it the --with-threads. maybe something is going wrong i'm not sure
> as to what to look out for?
> 
> Sorry about the lack of 'proof' so to speak. there is a lot of stuff in
> the output of configure, and all the makes.. if you can give me any
> advice as to what i should show you please tell me! i found this bit in
> the configure script.
> 
> Any ideas/comments/suggestions as to what's happening here? I really
> needs threads right now.
> ...
> checking for pthread.h... (cached) yes
> ...
> checking for --with-dec-threads... no
> checking for --with-threads... yes
> checking for --with-thread... yes
> checking for mach/cthreads.h... (cached) no
> checking for pthread_detach... (cached) no
> checking for kernel/OS.h... (cached) no
> checking for pthread_create in -lpthreads... (cached) no
> checking for pthread_create in -lpthread... (cached) no
> checking for pthread_create in -lc_r... (cached) no
> checking for __d6_pthread_create in -lthread... (cached) no
> checking for pthread_create in -lcma... (cached) no
> checking for usconfig in -lmpc... (cached) no
> checking for thr_create in -lthread... (cached) no
> ...

I think that you will need to do something like
	make distclean
to remove the existing built files and the current configuration.

-- 
Paul Duffin
DT/6000 Development	Email: pduffin at hursley.ibm.com
IBM UK Laboratories Ltd., Hursley Park nr. Winchester
Internal: 7-246880	International: +44 1962-816880



From a.eyre at optichrome.com  Wed Aug 11 04:52:39 1999
From: a.eyre at optichrome.com (Adrian Eyre)
Date: Wed, 11 Aug 1999 08:52:39 GMT
Subject: __init__ question
In-Reply-To: <Pine.LNX.4.10.9908100918190.10250-100000@starship.python.net>
References: <Pine.LNX.4.10.9908100918190.10250-100000@starship.python.net>
Message-ID: <001201bee3d6$dda83f10$3acbd9c2@peridot.optichrome.com>

> class Vector:
> 
> 	def __init__(self, x=0, y=0, z=0):
> 		self.x, self.y, self.z = x, y, z
> 
> 
> all examples from the original post, plus my own, work with the added
> benefit Python does all argument parsing for us.

True, but I got the impression from the original post that the number
of allowed arguments may be unlimited.




From cjw at connection.com  Tue Aug 31 08:31:55 1999
From: cjw at connection.com (Colin J. Williams)
Date: Tue, 31 Aug 1999 08:31:55 -0400
Subject: Bug in Tkinter?
References: <c75zoz849tq.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <37CBCB3B.6921D05D@connection.com>

Tim,

I get the same symptom with Win95.

Colin W.

tre17 at student.canterbury.ac.nz wrote:

> Have a look at the following code.  It should produce a window
> containing two radio buttons, one of which is active when the program
> starts.  But it does not!  Now remove the comments, so that there is a
> permanent reference to StringVar instance, and now it works?
>
> This is running Python 1.5.2 from the Debian package `python-tk 1.5.2-5'
> under Debian 2.2 (potato).
>
> =====================
> #!/usr/bin/python
>
> from Tkinter import *
> import sys
>
> class Foo(Frame):
>     def __init__(self, master):
>         Frame.__init__(self, master, relief='raised', bd=1)
>         mode = StringVar()
>         self.b1 = Radiobutton(self, text='Foo',
>                               variable=mode, value='foo',
>                               command=self.foo)
>         self.b2 = Radiobutton(self, text='Bar',
>                               variable=mode, value='bar',
>                               command=self.bar)
>         self.b1.pack(side='left', fill='x', expand=1)
>         self.b2.pack(side='right', fill='x', expand=1)
>         mode.set('bar')
>         print sys.getrefcount(mode)
>         #d['mode'] = mode
>         #print sys.getrefcount(mode)
>
>     def foo(self):
>         print 'foo'
>
>     def bar(self):
>         print 'bar'
>
> root = Tk()
> d = {}
> f = Foo(root)
> f.pack()
> root.mainloop()
> =====================
>
> Any comments?
>
> --
> Tim Evans






From stephan at pcrm.win.tue.nl  Tue Aug 31 09:42:54 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 31 Aug 1999 15:42:54 +0200
Subject: Newbie (read: pot. stupid) question on objects/namespaces
References: <8E339DC2FPaCmAnRDLM@news.dial.oleane.com>
Message-ID: <m3iu5w5an5.fsf@pcrm.win.tue.nl>

fredp at multimania.com.nospam (Fred Pacquier) writes:

> Hi,
> 
> I tried looking this up in the FAQ/tutorial and failed, but maybe it's a 
> terminology problem.
> 
> I'd like to know if it is possible, given the name of a "data attribute", 
> to find out the name of its "owner", ie the instance (namespace) where it 
> has been _created_ (the instance, not the class). This is probably unclear, 
> so to illustrate :
> 
> def func(a_thing,...)
>     	...
> 
> class foo:
>     	self.thing = ...
> 
> class bar:
>     	self.a_foo = foo()
>     	func(self.a_foo.thing,...)
> 
> Here, 'func' (might be anywhere, top-level, another class, in the same 
> module if it helps...) receives the data attribute 'thing' of instance 'a-
> foo' of class 'foo'. This instance was created (by ? inside ?) class bar, 
> or (probably) an instance 'a_bar' of class 'bar'.
> 
> The question is: given the simple call above with only the name as a 
> parameter, is there a (legal) way for function 'func' to determine that 
> 'a_thing' actually (lives in ? belongs to ?) 'bar' (or 'a_bar') ?

Let me try to rephrase the question first:

You want to know if it is possible to find out, given a class or
instance variable, to get at the class or instance.

Short answer: "no".

Longer answer: 
This is really a tricky problem what "class or instance variable" actually 
means in Python. If you have a class like this:

class Foo:
    x = 1

, then Foo.x refers to the number 1, which is an object in Python.
Many other classes Bar, Baz, and instances of those, might have references
to the same object 1. So 1 doesn't really have a single "owner".

Now, you could have a dedicated object which registers its official
"owner". Say you do that in this way:

class MyClass:
    def __init__(self, owner):
        self.owner = owner

You create instances of this object like this:

class Bar:
    x = MyClass(Bar)

, or, if you want to have the instance, rather than the class, as owner:

class Baz:
    def __init__(self):
        self.x = MyClass(self)

Note that in *both* cases you create a circular reference which means
that CPython cannot reclaim the class Bar or the instance of Baz.

Greetings,

Stephan 



From skip at mojam.com  Wed Aug  4 20:58:49 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed,  4 Aug 1999 19:58:49 -0500 (CDT)
Subject: Printing from python on Windows platform
In-Reply-To: <7oakee$9e2$1@m2.c2.telstra-mm.net.au>
References: <C12567C3.003009B3.00@derumg01.cyberlink.eds.com>
	<7oakee$9e2$1@m2.c2.telstra-mm.net.au>
Message-ID: <14248.57695.330650.399604@dolphin.mojam.com>

    Mark> Either way, what I can test is that you can specify the name of
    Mark> the printer as a network resource.  Eg:

    Mark> open("\\\\boots\\officejet", "w")

    Mark> will allow me to open my printer and write to it.

Dumb question from a non-Windows person: shouldn't that flag arg be "wb"?
We Unix weenies are always getting beat up by the Windows weenies for not
including the "b"inary flag when opening files.  I figured turn about is
fair play. ;-)

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From m.faassen at vet.uu.nl  Tue Aug 10 04:52:31 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 10 Aug 1999 08:52:31 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <19990805094020.A1838753@vislab.epa.gov> <37aacdf9.324492916@news.oh.verio.com> <934117874snz@vision25.demon.co.uk> <7om59k$cqc$1@newshost.accu.uu.nl> <934237862snz@vision25.demon.co.uk>
Message-ID: <7oop8f$hre$1@newshost.accu.uu.nl>

Phil Hunt <philh at vision25.demon.co.uk> wrote:
> In article <7om59k$cqc$1 at newshost.accu.uu.nl>
>            m.faassen at vet.uu.nl "Martijn Faassen" writes:

>> Phil Hunt <philh at vision25.demon.co.uk> wrote:
>> > I'm currently writing a generic GUI builder.
>> 
>> > It doesn't (and won't) have a graphical user interface, the user
>> > will write a user interface definition file, which will then be
>> > compiled by my tool into the programming language and GUI toolkit of
>> > your choice. A definition file might look like this:
>> 
>> [snip definition file example]
>> 
>> Why not use XML for this?

> IMO XML is too verbose. I want to be able to say:

>    textfield @Name "initial text"

> instead of:

>    <textfield id="Name" text="initial text"/>

> It is important to me that the tool is easy to use. This is because
> I intend to use it myself!

I see, that makes sense, but you do lose all the work that's being put into
XML, and all the hype, and standard Python libraries to deal with XML, etc.

Perhaps you can make it so that your abbreviated format can be readily
translated to XML?

> Also, IIRC XML is restrictive where it allows comments -- you can't put
> them just anywhere. My GUI builder will allow C++-style comments.

I'm not sure exactly on the XML rules for comments, but you can put them
before and after tags, I'm sure, which is quite a lot already. But again,
a preprocessor that turns your format into XML could work here.

> It will also use the C preprocessor, so for example one could
> write a macro for a label and associated textfield:

> #define labelTF(txt,id,size) \
>    label @id##Label txt textField @id cols=size
>  

You'd need a separate XML preprocessor to do something like this (or 
again just use XML for the backend). I'll stop harping on the XML for now,
I was just wondering; it would make it easier for people to pick up this
GUI definition language, and it seems applicable.

Regards,

Martijn




From thantos at brimstone.mecha  Sat Aug 21 21:16:49 1999
From: thantos at brimstone.mecha (Alexander Williams)
Date: Sun, 22 Aug 1999 01:16:49 GMT
Subject: carpe linguam
References: <199908201554.LAA11430@python.org> <37BDAD0A.FAFC980A@lineone.net>  <37bf4ba1@cs.colorado.edu>
Message-ID: <slrn7rujlr.34d.thantos@brimstone.mecha>

On 21 Aug 1999 19:00:17 -0700, Tom Christiansen <tchrist at mox.perl.com> wrote:
>to the collective you for such an unpleasant experience.    It's not
>the language.  It's the assholes.

I find this vaguely amusing, given the generally helpful and generous
attitude of cl.pythoneers in response to Tom's combative attitude;
this response is akin to going into a field, kicking visciously at an
ant bed, getting bitten a few times (and only so little because the
ants are lethargic) and then saying, "Hey, this kicking anthills
business would be fine if it weren't for the damn ants!"

If you want to use the language, Mr Christianson, use the language,
but don't try to make some moral argument of it.

(Why do I get the feeling he'll be going back to the Perl newsfroups
with the overblown, "Well, I /tried/ to extend a friendly hand, but
those damn ants just kept biting!"  Ants don't like the taste of boot,
in general.  Lesson for the learned.)

-- 
Alexander Williams (thantos at gw.total-web.net)
"In the end ... Oblivion Always Wins."



From khassan at aster.com.pk  Tue Aug 10 09:47:26 1999
From: khassan at aster.com.pk (khassan at aster.com.pk)
Date: Tue, 10 Aug 1999 13:47:26 GMT
Subject: Python says bad magic number
Message-ID: <7opaha$aij$1@nnrp1.deja.com>

Hi,

I am having trouble using python with database modules. I have a simple
script called "test" as follows:

#!/usr/local/bin/python

import gdbm
db = gdbm.open('detail', 'r')
print "There are", len(db), "keys in the database"
for key in db.keys():
        print key, db[key]

I get errors if I run the script as follows:

#python test
Traceback (innermost last):
  File "test", line 4, in ?
    db = gdbm.open('/usr/src/radiusContext-1.54/temp/detail', 'r')
gdbm.error: Bad magic number

What does this mean? I am using Linux 2.0.36, libc 5.4.46, python
1.5.2, gdbm 1.8.0, db 1.85.4. The detail file is actually the logfile
produced by cistron radius program (it is in livingston format).

Thanks,
Khurram,
khassan at aster.com.pk


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From thomas at xs4all.nl  Mon Aug 30 06:31:57 1999
From: thomas at xs4all.nl (Thomas Wouters)
Date: Mon, 30 Aug 1999 12:31:57 +0200
Subject: How to change default behaviour of a class?
In-Reply-To: <37CA5E90.28D60E87@xsoft.co.at>
References: <37CA5E90.28D60E87@xsoft.co.at>
Message-ID: <19990830123156.B29961@xs4all.nl>

On Mon, Aug 30, 1999 at 12:36:00PM +0200, Gerhard W. Gruber wrote:

> How can I change what is printed when I use the statement "print class"?
> Usually this leads to an output like <class instance at 9aad60> but I
> rather want it to print variables contained in my class instead of the
> pointer adress.

define a __repr__() method, that takes one argument (self):

>>> class A:
...     pass
... 
>>> class B:
...     def __repr__(self):
...             return "Cheesecake"
... 
>>> a, b = A(), B()
>>> print a
<__main__.A instance at 809a388>
>>> print b
Cheesecake
>>> 


-- 
Thomas Wouters <thomas at xs4all.net>

Hi! I'm a .signature virus! copy me into your .signature file to help me spread!



From news at dorb.com  Thu Aug 26 16:02:54 1999
From: news at dorb.com (Darrell)
Date: Thu, 26 Aug 1999 16:02:54 -0400
Subject: Fuzzy string matching?
References: <199908261746.TAA25440@dionysus.fw.cuci.nl>
Message-ID: <H3hx3.1648$v6.107145@typhoon.nyroc.rr.com>

http://www.sfgoth.com/~littlec/Soundex/info.html
http://www.lanw.com/java/phonetic/
--
--Darrell





From fredrik at pythonware.com  Mon Aug 23 09:00:34 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon, 23 Aug 1999 13:00:34 GMT
Subject: Date Calculations
References: <37BF5D1A.9865CDF@fnmail.com>
Message-ID: <00e001beed67$7e0ba7a0$f29b12c2@secret.pythonware.com>

Per Kistler <kistler at fnmail.com> wrote:
> I'm making some Date Calculaton class with things like
> deltaDays, addDays, daysInYear and so on, above the calendar 
> and time classes. After some 300 lines I thought, I better
> ask, whether that has already been done by someone else.

in addition to the other modules mentioned in this
thread, there's also something called:

    Demo/classes/Dates.py

in the standard distribution. here's the 'abstract':

# Date(month,day,year) returns a Date object.  An instance prints as,
# e.g., 'Mon 16 Aug 1993'.
#
# Addition, subtraction, comparison operators, min, max, and sorting
# all work as expected for date objects:  int+date or date+int returns
# the date `int' days from `date'; date+date raises an exception;
# date-int returns the date `int' days before `date'; date2-date1 returns
# an integer, the number of days from date1 to date2; int-date raises an
# exception; date1 < date2 is true iff date1 occurs before date2 (&
# similarly for other comparisons); min(date1,date2) is the earlier of
# the two dates and max(date1,date2) the later; and date objects can be
# used as dictionary keys.
#
# Date objects support one visible method, date.weekday().  This returns
# the day of the week the date falls on, as a string.
#
# Date objects also have 4 read-only data attributes:
#   .month  in 1..12
#   .day    in 1..31
#   .year   int or long int
#   .ord    the ordinal of the date relative to an arbitrary staring point
#
# The Dates module also supplies function today(), which returns the
# current date as a date object.

(it's written by timbot, btw).

</F>




From neelk at brick.cswv.com  Thu Aug 19 20:40:04 1999
From: neelk at brick.cswv.com (Neel Krishnaswami)
Date: 20 Aug 1999 00:40:04 GMT
Subject: RFC: Viper:  yet another python implementation
References: <37b075ae.2367293988@news.triode.net.au> 	<m3vham44g5.fsf@atrus.jesus.cam.ac.uk> 	<14257.54137.72005.705719@anthem.cnri.reston.va.us> <37b6f1c2.347828751@news.triode.net.au>  <37B82032.E7A11398@starmedia.net>
Message-ID: <slrn7rpbk8.7pl.neelk@brick.cswv.com>

On Mon, 16 Aug 1999 14:29:06 GMT, Andrew Csillag <andrew at starmedia.net> wrote:
>
>Also, you may want to look at how the SmallEiffel compiler does it's
>type inferencing...  They did it well enough to get rid of virtual
>function tables...  There's a few papers that describe what they did and
>the essence of how it works and the rather amazing perfomance numbers
>(and obviously the souce code is available to give you an implementation
>to look at -- written in Eiffel).
>
>http://SmallEiffel.loria.fr/papers/papers.html

This is well-worth reading, but the Eiffel folks were able to do their
optimizations because Eiffel is so much more static than Python -- for
example, classes can't be dynamically created in Eiffel. This is not
the case in Python, and I get the impression that John Skaller is
trying to use whole-program analysis to get around that a bit.

In that case, the Cecil papers:

    http://www.cs.washington.edu/research/projects/cecil/papers

are the best place to look for ideas. Cecil is a dynamic OO language
that Craig Chambers has been using to work out techniques for wringing
speed out of dynamic languages. The only danger is that it will
inspire John to add predicate methods to Python. :)


Neel



From ngps at madcap.dyn.ml.org  Sat Aug 21 13:43:34 1999
From: ngps at madcap.dyn.ml.org (Ng Pheng Siong)
Date: 21 Aug 99 17:43:34 GMT
Subject: [ANNOUNCE] M2Crypto 0.01
References: <37bd6fa4.0@news.cyberway.com.sg> <37BED351.55E7DA98@lemburg.com>
Message-ID: <37bee546.0@news.cyberway.com.sg>

<sheepish>
Ahem, I was fiddling with my newsreader when I posted this,
and I thought the posting didn't make it out. Since it did, 
and I tried multiple times, I think there will be multiple
announcements. 
</sheepish>

According to M.-A. Lemburg <mal at lemburg.com>:
> Ng Pheng Siong wrote:
> > I am pleased to announce the release of M2Crypto 0.01;
> > the package is available at http://www.post1.com/home/ngps
> 
> Could you please tell me where that server is located and whether
> it is legal to download the software from Germany ? 

The server is located in Singapore. AFAIK, there is no law against
export of cryptography from Singapore.


> Anyway, mxCrypto development will continue to eventually include
> most APIs of OpenSSL.

Hey, there are umpteen Java crypto libraries. Python can live with
three or five. ;-)

Cheers.

-- 
Ng Pheng Siong <ngps at post1.com>




From sphilip at t2.technion.ac.il  Sat Aug  7 08:09:59 1999
From: sphilip at t2.technion.ac.il (Philip Derbeko)
Date: Sat, 07 Aug 1999 12:09:59 +0000
Subject: Help. Tracking ButtonRelease events
Message-ID: <37AC2217.F69CD3D6@t2.technion.ac.il>

Hello.

I'm writing an application part of which is a GUI to draw graphs.
What i want is create verticies and edges between them,
i create vertex by clicking on Button 1 ( taken from Demo/Tkinter/matt
scripts )
and edge created by clicking on Button 3 and dragging the mouse to the
destination vertex.
In order to do it every vertex  binded events ButtonPress-3 and
ButtonRelease-3 ,
the problem i run into is that when i press button 3 and drag it to the
destination vertex,
source vertex gets the ButtonRelease-3 event and not destination.

Am i missing something and how can it be solved ?

Thanks in advance
Philip





From earlybird at mop.no  Mon Aug 16 07:28:26 1999
From: earlybird at mop.no (Alexander Staubo)
Date: Mon, 16 Aug 1999 13:28:26 +0200
Subject: 16-bit Python -- does it exist?
Message-ID: <MPG.122214458f7ab0d2989685@news.mop.no>

Subject says all. A friend (really!) would like to run Python on an old 
80286. Possible?

-- 
Alexander Staubo             http://www.mop.no/~alex/
"QED?" said Russell.
"It's Latin," said Morgan. "It means, 'So there you bastard'."
--Robert Rankin, _Nostramadus Ate My Hamster_



From jstevens at basho.fc.hp.com  Fri Aug 13 23:04:39 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:04:39 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132310.RAA06953@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:10 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ceb7@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :list.append( newElement )
> :
> :to Perl's closest equivalent:
> :
> :list[ $#list + 1 ] = $newElement
> 
> Don't be silly.  That's merely
> 
>     push @array, $new_element;
> 
> Although you can use the other way if you'd like.
> I hope that freedom doesn't bug you too much.

However, using a stack operation on an array that you are currently
attempting to use to represent a list provides many opportunities
for misunderstandings, and out right bugs.

John S.




From pehr at alum.mit.edu  Sun Aug  8 17:44:22 1999
From: pehr at alum.mit.edu (pehr anderson)
Date: Sun, 08 Aug 1999 17:44:22 -0400
Subject: how do I?
References: <19990807033959.79942.qmail@hotmail.com>
Message-ID: <37ADFA36.237FB87@alum.mit.edu>

Dear Jake,
You may find the tkinter doucmentation you need at the following URL:
http://www.pythonware.com/library.html

I'm doing a little tkinter stuff but haven't yet found a complete reference.
    -pehr

Jake Ramirez wrote:

> Hello out there,
> I am quite new at using python and Tcl/tk scripting tools. How do I make
> them work together. Can I get a solid web reference on that? Thanks for the
> info
>
> Jakester
>
> _______________________________________________________________
> Get Free Email and Do More On The Web. Visit http://www.msn.com




From zwi at turboimage.com  Mon Aug 23 05:27:22 1999
From: zwi at turboimage.com (Zwika Moshkowzki)
Date: Mon, 23 Aug 1999 11:27:22 +0200
Subject: Q: Authentication with httplib
Message-ID: <7pr09c$dag$1@news.netvision.net.il>

Hi,

I am using httplib to access a Page on a Server:

    .....
     h = httplib.HTTP(Server)
     h.putrequest('GET', Page)
     h.endheaders()
    .....

When Server is configured not to require a passwd, I set Server="myServer",
Page="/myPage",
and everything is OK.

My Q regards the situation where there is a need for authentication: Where
should I put the Username+Passwd?

Thanks in advance!

zwikaM





From elephant at squirrelgroup.com  Thu Aug 19 12:07:35 1999
From: elephant at squirrelgroup.com (elephant)
Date: Fri, 20 Aug 1999 02:07:35 +1000
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <7ph3p8$7ci@junior.apk.net>
Message-ID: <MPG.1226d6d63a058d69989c5c@news-server>

Mark W. Schumann writes ..
>In article <37b62939 at cs.colorado.edu>,
>Tom Christiansen  <tchrist at mox.perl.com> wrote:
>>Gosh yes.  I think it was the rogue source that had something like:
>>
>>    #define until(expr) while(!expr)
>>    #define otherwise   break; default:
>>
>>I don't remember whether it also had these, but I have
>>seen them elsewhere:
>>
>>    #define forever()  for(;;)
>>    #define unless(expr) if(!expr)
>
>I've seen
>
>#define ARBEGIN {
>#define AREND   }

in his section on Privacy in "Advanced Perl Programming" Sriram 
Srinivasan mentions (as a counterexample to syntactically enforced 
privacy) seeing the following in some C++ code just before a header file 
was included

  #define private public

there must be something wrong with me .. because I'd never thought of 
that

-- 
 jason - elephant at squirrelgroup.com -



From gmcm at hypernet.com  Mon Aug 23 23:53:46 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 23 Aug 1999 22:53:46 -0500
Subject: "def" vs "sub" (was RE: More random python observations from
In-Reply-To: <37C1EBBA.573DA45C@starmedia.net>
Message-ID: <1276659530-18221306@hypernet.com>

Andrew Csillag wrote:

> "Juergen A. Erhard" wrote:

> > And could you tell us where this mythical PyLint is?  

>... it's a creation of the venerable Andrew Kuchling as part of
> the kwParsing package (a parser generator package), which also
> includes the gadfly relational database - way cool.
> 
> http://www.chordate.com/kwParsing/index.html

Andrew is venerable, and you have the right url, but Aaron 
Watters is the author.

- Gordon



From bwarsaw at cnri.reston.va.us  Fri Aug 27 12:17:26 1999
From: bwarsaw at cnri.reston.va.us (Barry A. Warsaw)
Date: Fri, 27 Aug 1999 12:17:26 -0400 (EDT)
Subject: Library philosophy
References: <37C57A80.D5C62704@prescod.net>
	<02a101bef069$f1ecc190$f29b12c2@secret.pythonware.com>
Message-ID: <14278.47638.482198.49815@anthem.cnri.reston.va.us>

>>>>> "FL" == Fredrik Lundh <fredrik at pythonware.com> writes:

    FL> on the other hand, it wouldn't surprise me if
    FL> Barry and Fred has already cleaned up the 1.6
    FL> library, while Guido was out of town... ;-)

We have, but don't tell Guido.  It's in the CVS tree, and you just
need to do the following:

    cvs up -r"keep_quiet_or_we_will_get_a_gumball_shoved_down_our_throats"

-Barry



From greg at cosc.canterbury.ac.nz  Mon Aug  9 17:55:39 1999
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Mon, 09 Aug 99 21:55:39 GMT
Subject: Lexical Scoping Patch
Message-ID: <mt2.0-7542-934235739@news.informatik.uni-bonn.de>

I think I've found a way of adding lexical scoping to
Python while avoiding the worst of the cyclic reference
problems.

To find out more, and download the patch, see:

http://www.cosc.canterbury.ac.nz/~greg/python/lexscope.html

== 
Greg Ewing, Computer Science Dept, | The address below is not spam-
University of Canterbury,	   | protected, so as not to waste
Christchurch, New Zealand	   | the time of Guido van Rossum.
greg at cosc.canterbury.ac.nz



From ionescu at pik-potsdam.de  Thu Aug 19 02:53:58 1999
From: ionescu at pik-potsdam.de (Cezar Ionescu)
Date: Thu, 19 Aug 1999 08:53:58 +0200
Subject: The stupidest question ever seen in this newsgroup...
References: <37BB8F9A.6C204793@rocler.qc.ca>
Message-ID: <37BBAA06.A26AC887@pik-potsdam.de>

The program is supposed to factor y, so you have to provide it with a
value for y. Just write a line that initializes y, e.g.
> 
> #!/usr/bin/python
> 

y = 24

> x = y / 2
> while x > 1:
>     if y % x == 0:
>          print y, 'has factor', x
>          break
>     x = x - 1
> else:
>      print y, 'is prime'
> 

Good luck,
Cezar.



From Svante.Kleist at NEMESIS.se  Sat Aug 28 04:00:05 1999
From: Svante.Kleist at NEMESIS.se (Svante Kleist)
Date: Sat, 28 Aug 1999 10:00:05 +0200
Subject: IDLE: What to use instead of sys.argv[0] ?
Message-ID: <MPG.1231b572a3fb6029989682@news.pi.se>

Dear Pythoneers,

>From within a Python script executed in IDLE, how can I get
its path/file? sys.argv[0] gives me this:

	E:\local\Python\Tools\idle\idle.pyw

Thanks for any tip.

/ Svante Kleist - NEMESIS systemDesign, Stockholm



From stephen at tocol.chc.us.ac.com  Fri Aug 27 13:24:08 1999
From: stephen at tocol.chc.us.ac.com (Stephen M. Shelly)
Date: 27 Aug 1999 12:24:08 -0500
Subject: how to get output from system call into data object
Message-ID: <863dx5gmrr.fsf@tocol.chc.us.ac.com>

I am still VERY new to python, and I have a pretty ugly logical 
problem to solve. I have tried using perl, but it is very nasty 
looking when I try to use multi-dimensional arrays. I need to run
a command which will return data in column format, I want to extract 
particular columns into data structures that I can then manipulate.
I currently have:
import os
system("my_command | awk {'print $What $I $ Need'}")
 i would like to fill a list of objects with this output, but I 
do not know how......

I know this is not what I should be doing, but I do not have a 
good enough understanding of python to figure out how to create 
my list of data objects 
from this output.

is it like:
MyList = system("Command | awk {print what i need}") ??????????

As always, any and all assistance is rewarded in the afterlife as 
well as with my deepest gratitude.
PEACE 



From tchrist at mox.perl.com  Thu Aug 19 15:10:06 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 12:10:06 -0700
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu> <slrn7ro9ns.h6v.jblaine@shell2.shore.net> <37bc2ee5@cs.colorado.edu> <7phfe0$19so$1@nntp6.u.washington.edu>
Message-ID: <37bc487e@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Donn Cave <donn at u.washington.edu> writes:
:Searchable CLI documentation is a good thing, and I take that as the main
:point of your post here, but the way Perl and Tcl dump man pages into man3
:isn't something Python needs to emulate on my account.  

I agree entirely, and Perl shouldn't be doing that.  It uses 
it's only manpath, which is separate.  For example:

    % perl -V:man.dir
    man1dir='/usr/local/perl/man/man1'
    man3dir='/usr/local/perl/man/man3'

That tells me that I can just use as an alias pman to be
"man -P /usr/local/perl/man" or "MANPATH=/usr/local/perl/man man"

Your site may have them installed in different locations, but
the principle is the same.  

But this doesn't help the module documentation issue.  When a 
new module is installed, I get to see it with apropos (that's 
"man -k" for some of you):

% pman -k net
perlfaq9 (1)         - Networking ($Revision: 1.24 $, f(CW$Date: 1999/01/08 05:39:48 $)
lwp-download (1)     - fetch large files from the net
CGI::Cookie (3pm)    - Interface to Netscape Cookies
IO::Socket::INET (3pm) - Object interface for AF_INET domain sockets
Mail::Internet (3pm) - manipulate Internet format (RFC 822) mail messages
Net::Cmd (3pm)       - Network Command class (as used by FTP, SMTP etc)
Net::Domain (3pm)    - Attempt to evaluate the current host's internet name and domain
Net::DummyInetd (3pm) - A dummy Inetd server
Net::FTP (3pm)       - FTP Client class
Net::NNTP (3pm)      - NNTP Client class
Net::Netrc (3pm)     - OO interface to users netrc file
Net::PH (3pm)        - CCSO Nameserver Client class
Net::POP3 (3pm)      - Post Office Protocol 3 Client class (RFC1081)
Net::Ping (3pm)      - check a remote host for reachability
Net::SMTP (3pm)      - Simple Mail Transfer Protocol Client
Net::SNPP (3pm)      - Simple Network Pager Protocol Client
Net::Telnet (3pm)    - interact with TELNET port or other TCP ports
Net::Time (3pm)      - time and daytime network client interface
Net::hostent (3pm)   - by-name interface to Perl's built-in gethost*() functions
Net::netent (3pm)    - by-name interface to Perl's built-in getnet*() functions
Net::protoent (3pm)  - by-name interface to Perl's built-in getproto*() functions
Net::servent (3pm)   - by-name interface to Perl's built-in getserv*() functions
Socket, sockaddr_in, sockaddr_un, inet_aton, inet_ntoa (3pm) - load the C socket.h defines and structure manipulators

Nifty, eh?  I wonder how to do that when you install modules in Python?
How close an automated, MakeMaker-like thing do you have?

:Man pages could be disposed in the same name-space-sensitive way we
:design libraries, modules, etc., and if anyone feels like working the
:Python documentation into man pages I hope that will be considered.

I'm still dumb-founded that so many Python people don't use the manpages,
or evens expects them!  Considering how much I've complained, I'd
certainly be willing to give it a try.  Do they come in a basic format
that can just be passed through a filter?  And no, we don't really use
manpages for Perl.  We use pod, then translate to many formats. 
The Prisoners of Bill gets HTML, regular Unix users get manpages,
and anyone can have simple text.  And everyone is happy.

--tom
-- 
 If you choose not to decide, you still have made a choice.  --Rush



From tim_one at email.msn.com  Tue Aug 10 23:23:47 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 10 Aug 1999 23:23:47 -0400
Subject: Getting at line numbers with parsermodule
In-Reply-To: <14256.39346.878530.327463@buffalo.fnal.gov>
Message-ID: <000001bee3a8$ed930a00$ca2d2399@tim>

[Charles G Waldman]
> ...
> I'm trying to use the parser module.
> ...
> I have trouble printing out the line number in the source where
> the offending code occurred.

RTFM for this one:  ast2list and ast2tuple each take an optional line_info
argument, defaulting to false.

> I tried to simply go throught the AST object (converted to a tuple) and
> look for NEWLINE tokens, count these as I go throught the AST to get a
> current line-count.  This doesn't seem to work correctly since there are
> fewer NEWLINE tokens than there were lines in the original file.

Right, a NEWLINE in Python's grammar occurs at the end of a statement, and
not every physical \n ends a statement.  For example, you won't get NEWLINE
at the end of any line that's continued (via string, backslash, or
open-brace/bracket/paren continuation).  So far as Python's parsing is
concerned the statement hasn't ended yet.

> What I'd like to be able to do is, given any node in an AST object
> returned by compile.suite, is to know the line number in the input
> that corresponds to this node.  Is this doable?

No.  Specifying a true line_info argument will get you line numbers for
terminal tokens, though.

would-the-manual-lie?-ly y'rs  - tim





From JamesL at Lugoj.Com  Thu Aug 19 00:44:11 1999
From: JamesL at Lugoj.Com (James Logajan)
Date: Wed, 18 Aug 1999 21:44:11 -0700
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org> <00c301bee884$42b1d3b0$f29b12c2@secret.pythonware.com>
Message-ID: <37BB8B9B.6092BB93@Lugoj.Com>

Fredrik Lundh wrote:
> This module [os] provides a more portable way of using
> operating system (OS) dependent functionality than
> importing an OS dependent built-in module like posix
> or nt.

POSIX is OS dependent? Well, yeah, I guess in a self-referential sort of
way. Non-portable? Yeah, I guess in an inevitable sort of way any standard
must stand apart from all other standards in its realm. I think there is wry
bit of humor in the above quoted text when it is contrasted with the
expanded acronym:

"FIPS 151-2, POSIX: Portable Operating System Interface for Computer
Environments"

I guess if the authors of POSIX were looking for a mature OS that was itself
known to be portable across machine platforms to use as their template, they
could have used Windows NT. Fortunately they didn't have a time machine
handy. (Only MS would fix something that wasn't broken. But when there is a
world to conquer.... I vaguely recall once hearing Bill Gates saying Unix
was going to be the OS of the future. This was back when MS pushed Xenix as
THE robust OS; before DOS really came on the scene.)



From arcege at shore.net  Tue Aug 17 18:33:07 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Tue, 17 Aug 1999 22:33:07 GMT
Subject: Error Message: Can't assign to Operator
References: <Pine.LNX.4.04.9908172133450.24702-100000@natal.stamp-coin.com>
Message-ID: <Dqlu3.436$mV.157066@news.shore.net>

Jonathon <jblake at stamp-coin.com> wrote:

: I keep hitting this error message:
: SyntaxError:  can't assign to operator(line 3056)
: 
: Line 3056 is
: meta-previous = string.join( the_ultimate_file_destination_is_this, string_record_id, suffix )
: 
: the_ultimate_file_destination_is_this, string_record_id and  suffix 
: are all defined elsewhere.

: Questions:

: Why does that construct not work?
: How can I fix this construction?

It is probably because "meta-previous" is parsed as "meta" "-" "previous"
a subtraction expression.  Change the minus ("-") to an underscore ("_")
and it should work better.  I've done this. :)

  -Arcege




From tim_one at email.msn.com  Tue Aug 17 21:20:21 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 21:20:21 -0400
Subject: Unsorting(randomizing) a sequence
In-Reply-To: <14265.49248.386119.227023@bitdiddle.cnri.reston.va.us>
Message-ID: <000601bee917$d6fa4c80$eaa22299@tim>

[Jeremy Hylton]
> ...
> The "shufflesort" or "Tim Peters" method, which I would offer as
> preferred names (because I don't particularly want my name associated
> with it and Tim's name is as good as anyone's),

Indeed, I'm *thrilled* to be associated with this worthless method, Jeremy!
I must have it inscribed on my tombstone.

> depends an awful lot on the Python quicksort implementation (another
> good reason to call it the Tim Peter's method <wink>).

Since Python no longer uses a quicksort, that's even more appropriate
<wink>.

[attempting to explain shufflesort's bad behavior]
> ...
> I think this follows from the quicksort approach.
> ...

It certainly depends on the details of the sort implementation.  For the
size of inputs you've been looking at, Python actually uses a pure binary
insertion sort after finding the longest already-sorted prefix, and the only
comparison outcome it looks at is the two-way "less than, or not?".  Your
choice function gives equal weights to -1, 0 and 1, the latter two of which
the sort takes as meaning "no".

For the two-element case, it will leave the input alone 75% of the time.
Now you'll have no problem completing an exact analysis for the general case
<wink>>

almost-done-with-the-N=1-case-myself-ly y'rs  - tim





From da at ski.org  Tue Aug  3 17:53:11 1999
From: da at ski.org (David Ascher)
Date: Tue, 3 Aug 1999 14:53:11 -0700 (Pacific Daylight Time)
Subject: FYI: CareTracker from LinuxCare
Message-ID: <Pine.WNT.4.04.9908031448080.145-100000@rigoletto.ski.org>

Looks like a bug-tracking system layered on top of jitterbug (they say
it's 'highly optimized'. Available as Open Source, and they host instances
for open source projects.

HTTP://www.linuxcare.com/developer/caretracker.html/001a63df96778889

and

HTTP://www.linuxcare.com/developer/opensource_projects.html/001a63df96778889

(absolutely no endorsment on my part, but I have to say I've been waiting
for someone to offer this service for a while, and their system seems very
responsive).

--david




From ivanlan at callware.com  Tue Aug 10 16:23:03 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Tue, 10 Aug 1999 20:23:03 GMT
Subject: posix.error is a tuple, not an object or a string as docs say
References: <yttzozzbm11.fsf@gilgamesh.cse.ucsc.edu>
		<KM_r3.1361$EG4.234560@news.shore.net> <yttu2q7bgb8.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <37B08A27.4020E16C@callware.com>

Hi Ben, All--

Ben Gertzfield wrote:
> 
> >>>>> "Michael" == Michael P Reilly <arcege at shore.net> writes:
> 
>     Michael> The long and short of this is that you can use it like a
>     Michael> tuple if more than one argument was given, and as the
>     Michael> object otherwise.
> 
>     Michael> For some of your own reassurance, print the type of
>     Michael> "error" in the except clause.
> 
> Right, but my problem is that I can't use it as an object as the
> documentation says..
> 
> Python 1.5.1 (#1, Dec 17 1998, 20:58:15)  [GCC 2.7.2.3] on linux2
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import os
> >>> try:
> ...     input = os.fdopen(42)
> ... except os.error, error:
> ...     print error.errno
> ...     print error.strerror
> ...
> Traceback (innermost last):
>   File "<stdin>", line 4, in ?
> AttributeError: errno
> 
> Yet the documentation says:
> 
> "When exceptions are classes, this exception carries two attributes,
> errno and strerror. The first holds the value of the C errno variable,
> and the latter holds the corresponding error message from strerror()."
> 
> So something else is a bit wacky, no?
> 

Try this script on a nonexistent file:  "python readit.py
doesntexist.txt", for example.
---------------------------------------
#!/usr/local/bin/python

import os
import sys
import errno

if __name__ == "__main__" :
    if len ( sys.argv ) > 1 :
	try:
	    f = open ( sys.argv[ 1 ], "rb" )
	except :
	    x = sys.exc_info()
	    print type (x)
	    print type(x[0])
	    print dir(x[0])
	    print type(x[1])
	    print dir(x[1])
	    print x[1].errno
	    print type(x[2])
	    print "No file named %s!" % ( sys.argv[ 1 ] )
	    sys.exit(1)
	while 1 :
	    t = f.readline ( )
	    if t == '' :
		break
	    if '\n' in t :
		t = t[ : -1 ]
	    if '\r' in t :
		t = t[ : -1 ]
	    sys.stdout.write ( t + '\n' )
	f.close ( )
------------------------------------------------


> --
> Brought to you by the letters C and H and the number 11.

<brought-to-you-by-chapter-8-of-teach-yourself-python-in-24-hours>-ly
y'rs,
Ivan;-)
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From nhead at houbits.com  Mon Aug 16 11:29:05 1999
From: nhead at houbits.com (Nigel Head)
Date: Mon, 16 Aug 1999 17:29:05 +0200
Subject: Difficulty configuring web server for Python
References: <7p9855$5tg$1@nnrp1.deja.com>
Message-ID: <37B82E40.BF51C2E0@houbits.com>

This sounds kind of like it my be an "iserver" setup,  right ?

They do a chroot() so that the path for the python executable should be
/usr/local/bin (ie. remove /usr/home/campusch) from the front of any paths
you use when running from under the web server.

I've fallen foul of this a few times - the telnet login shell isn't at all
representative of the environment scripts actually run in. My setup has a
"virtual" command which runs its arguments in such a chroot'ed environment
to allow testing from the command line.

Nigel.




From chega_ at yahoo.com  Fri Aug 13 01:24:21 1999
From: chega_ at yahoo.com (Vadim Chugunov)
Date: Thu, 12 Aug 1999 22:24:21 -0700
Subject: exec & function locals
Message-ID: <7p0a67$8su@news.dns.microsoft.com>

A question for those who are familiar with the Python's source code.

As well known,
    exec f.func_code in dictGlobals,dictLocals
will not update dictLocals because of the local binding optimizations
(LOAD_FASTs and STORE_FASTs)

However, I figure, this could be easily fixed by adding
    if (locals)
        PyFrame_FastToLocals(f);
at the end of eval_code2, just before the current frame is released.

Well, is there any particular reason, that I'm missing, for not having such code
in eval_code2 ?
Anyone ?

Vadim






From aahz at netcom.com  Wed Aug 11 09:44:06 1999
From: aahz at netcom.com (Aahz Maruch)
Date: 11 Aug 1999 13:44:06 GMT
Subject: sys.argv attribute error
References: <37B16C47.E6212B21@enews.nrl.navy.mil>
Message-ID: <7orun6$rij@dfw-ixnews16.ix.netcom.com>

In article <37B16C47.E6212B21 at enews.nrl.navy.mil>,
Zach  <wiener at enews.nrl.navy.mil> wrote:
>
>Within a python file, I am trying to pring sys.argv, but I get an
>AttributeError with argv.  I can print sys.path fine, however.  Does
>anyone know why I would be having problems with printing argv?  Thanks.

Are you printing sys.path on the line immediately before your attempt to
print sys.argv?
--
                      --- Aahz (@netcom.com)

Androgynous poly kinky vanilla queer het    <*>      http://www.rahul.net/aahz/
Hugs and backrubs -- I break Rule 6  (if you want to know, do some research)



From tbryan at arlut.utexas.edu  Tue Aug  3 08:20:33 1999
From: tbryan at arlut.utexas.edu (Thomas Bryan)
Date: Tue, 03 Aug 1999 07:20:33 -0500
Subject: Import a lot of modules, or, import 'foo'
References: <874silabe7.fsf@linuxcare.com> <1%ro3.4357$VL2.467111@news.direcpc.com> <871zdpa9lb.fsf@linuxcare.com> <37A27A54.A737AE62@vision.arc.nasa.gov> <7o5fb1$lat$1@ssauraac-i-1.production.compuserve.com>
Message-ID: <37A6DE91.F7ADB618@arlut.utexas.edu>

"T. C. Mits" wrote:

>     I see some great replies and extra info given on this NG.   Is someone
> moving this to a FAQ or resource or must one do searches all over the place
> for it?   If not is deja news the only way to search this NG?

As with most newsgroups, dejanews is the second place to look for info.
The FAQ is the first: http://www.python.org/doc/FAQ.html

All PSA members can submit/modify FAQ entries online.  I don't know
how often people actually modify the FAQ, but I generally only do it
to add a *frequently* asked question.  Otherwise, I just let dejanews
archive the thread.  If you'd really like to modify the FAQ yourself, 
join the PSA!  The membership year starts Oct. 1.

Since the FAQ is searchable, should PSA members make an effort to add 
lots of entries to the FAQ?  I'm always torn between the "all the info
in one place" and "so much info in one place that people drown"
philosophies.
Does anyone have a good rule of thumb for Python FAQ?

-------------------------------------------
Tom Bryan
Applied Research Laboratories
University of Texas at Austin



From jdnier at execpc.com  Wed Aug  4 14:28:26 1999
From: jdnier at execpc.com (David Niergarth)
Date: Wed, 4 Aug 1999 13:28:26 -0500
Subject: win32api -- long vs short filenames
Message-ID: <7oa0n0$72n@newsops.execpc.com>

I'm trying to expand a mangled DOS 8+3 name (from WinNT4.0) into the long
version of the name. There's a win32api function for doing the opposite, for
example,

>>> win32api.GetShortPathName(r'c:\long_path_name\long_file_name.txt')
'c:\\LONG_P~1\\LONG_F~1.TXT'

But I can't figure out how to expand a mangled name into the original long
name. Any ideas?

--David Niergarth






From mlh at idt.ntnu.no  Thu Aug 26 16:04:01 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 26 Aug 1999 22:04:01 +0200
Subject: Fuzzy string matching?
References: <199908261746.TAA25440@dionysus.fw.cuci.nl> <WXgx3.1646$v6.107045@typhoon.nyroc.rr.com>
Message-ID: <y0j4shmux5a.fsf@vier.idi.ntnu.no>

"Darrell" <news at dorb.com> writes:

> Your distance thing is cool.
> But check out soundex.
> These are a bit redundent.
> http://www.outfitters.com/genealogy/what-soundex.html
> http://www.darken.com/soundex.htm
> http://www.mayrand.org/soundex.html

I believe I mentioned soundex in my original post - but thanks for the
URLs.

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From Sven.Drescher at dlr.de  Tue Aug  3 11:12:25 1999
From: Sven.Drescher at dlr.de (Sven Drescher)
Date: Tue, 3 Aug 1999 17:12:25 +0200
Subject: Libraries needed, please help!!!
References: <7ns6o5$jo2@news.go.dlr.de><004e01beda98$d14f9390$f29b12c2@secret.pythonware.com><7o3moo$dse@news.go.dlr.de><14245.40279.79936.219772@weyr.cnri.reston.va.us><7o4ak3$imi@news.go.dlr.de><37A5B809.A29DDFF5@GermanLloyd.org><7o5vvh$lai@news.go.dlr.de> <14247.554.177824.770257@weyr.cnri.reston.va.us>
Message-ID: <7o703n$lcc@news.go.dlr.de>

>  > > Order matters? Try:
>  > >
>  > > CC -I/path/to/include/ -L/path/to/lib/ -o gui gui.cxx -lpython1.5
>  > >
>  > > Good luck
>  > >
>  > > Berthold
>  > > --
>  > > email: hoel at GermanLloyd.org
>  > I tried, but there are more undefined symbols yet. All symbols are
first
>  > referenced in libpython1.5.a(socketmodule.o)!
>
>   Are the missing symbols socket-related (gethostbyname, etc.)?  If
> you're doing all this on Solaris, you need to link with "-lsocket
> -lnsl" to get all the symbols.
>
>   -Fred
> --
> Fred L. Drake, Jr.      <fdrake at acm.org>
> Corporation for National Research Initiatives

Thanks!
Many errors are removed.

But there are two undefined symbols yet:
 - importdl.o
 - regexpr.o

Could you help me more, please???

Thanks a lot!
Sven






From steve_w at localhost.localdomain  Sun Aug 22 08:19:12 1999
From: steve_w at localhost.localdomain (steve_w at localhost.localdomain)
Date: Sun, 22 Aug 1999 12:19:12 GMT
Subject: What is a troll by the way? [offtopic]
References: <zelv3.13444$x04.816752@typ11.nn.bcandid.com> <37bdeb1b.31915768@news.erols.com> <37BDF3A7.75FEA04F@vision.arc.nasa.gov>
Message-ID: <slrn7rvqoj.3sd.steve_w@localhost.localdomain>

On Sat, 21 Aug 1999 00:32:39 +0000, Chad Netzer wrote:
>Robert Kern wrote:
>
>> _Troll_ is a USENET term for someone that deliberately posts an
>> inflammatory message simply to stir up argument.
>
>I've found it funny to watch this term evolve over the past few years.
>It started (I believe) as a reference to the fishing term, and used to
>mean "trolling for flames" (ie. hoping to bring flames to the
>discussion), perhaps employing "flame bait" for this purpose.
>
Actually, the first and best sort of trolling was extremely subtle.
The idea was to insert some subtle flaws into a posting, usually
in the midst of a flamewar, to elicit the automatic followups from
folks that are too fast with the followup button, in order to expose
this tendency.

Steve Wall
 
>
>



From ritm at gnet.tn  Sun Aug 29 17:26:25 1999
From: ritm at gnet.tn (MB)
Date: Sun, 29 Aug 1999 23:26:25 +0200
Subject: building Python+readline on Suse 6.0
Message-ID: <37C9A580.A3A3D95A@gnet.tn>

Hi
I have problems building python with realine +on Suse 6.0 so I found a
message posted on 1999/30/07 posted by Henning Shroeder about building
python with readline and curses. I made the sugggested changes (adding
-L/usr/lib/termcap) but I receive a warning an python doesn't work (only
when I make with readline)
The message::
-----------------
gcc -Xlinker -export-dynamic python.o \
                ../libpython1.5.a -L/usr/lib/termcap -lreadline
-ltermcap
                 -ldl   -lm  -o python
/usr/i486-linux/bin/ld: warning: libc.so.6, needed by
/usr/lib/termcap/libtermcap.so, may conflict with libc.5.so
-----------------

how to solve it ?

Thanks a lot

Mourad




From zorro at zipzap.ch  Fri Aug 13 11:21:28 1999
From: zorro at zipzap.ch (Boris Borcic)
Date: Fri, 13 Aug 1999 17:21:28 +0200
Subject: Does Python run under Windows 2000? (build 2031)
References: <37B40D58.DD6AADC0@RobotVision.com>
Message-ID: <37B437F8.D07EDEB8@zipzap.ch>

James E. Green wrote:
> 
> I'm running Windows 2000 RC1 (build 2031, the very latest).

2000 ? We aren't there yet. Apparently your OS suffers from Y2K.

BB



From jeffp at crusoe.net  Fri Aug 13 11:05:26 1999
From: jeffp at crusoe.net (Jeff Pinyan)
Date: Fri, 13 Aug 1999 11:05:26 -0400
Subject: Interesting thread on comp.lang.perl.misc
In-Reply-To: <v9Ws3.1713$EG4.332188@news.shore.net>
References: <37B42034.33F77AD5@strs.co.uk> <v9Ws3.1713$EG4.332188@news.shore.net>
Message-ID: <Pine.GSO.4.10.9908131047510.3761-100000@crusoe.crusoe.net>

On Aug 13, Michael P. Reilly blah blah blah:

> So what is the best forum?  Good question.  The best would be to find
> papers by reputable persons who fairly judge the languages.  You got a
> very good response from Tom C.  Read what he says; he's fair about both
> languages.

I program both Perl and Python.  I'm in clpmisc and clpython.  What do I
do when such debates come around?  I ignore them.  I do, however, be sure
to make well researched and as unbiased comments as I can if I do happen
to say a sentence or two.

Many people find python's whitespace dependency to be stupid.  I love it.

Many people find perl's leading symbol characters (@$%&*) to be stupid.  I
love them.

I wish Perl had Object.Attr OO functionality as in Python.

I wish I could change a string in-place with a function.*

But I like the two languages.  Holy wars are not for me.


* meaning:

def func(str):
  # something modifying str 

str = "foo"
func(str)
print str # prints... "bar", for example

In Perl, you can modify arguments to subroutines:

sub func { $_[0] = "bar" }
$str = "foo";
func($str);
print $str; # prints "bar"
func("jeff"); # that's an error, you can't change a constant

Ok.  That's it for me. :)

--
jeff pinyan    japhy at pobox.com  japhy+perl at pobox.com  japhy+crap at pobox.com
japhy's little hole in the (fire) wall:       http://www.pobox.com/~japhy/
japhy's perl supposit^Wrepository:       http://www.pobox.com/~japhy/perl/
The "CRAP" Project:                 http://www.pobox.com/~japhy/perl/crap/
CPAN ID: PINYAN           http://www.perl.com/CPAN/authors/id/P/PI/PINYAN/




From moshez at server.python.net  Sat Aug 28 13:00:42 1999
From: moshez at server.python.net (Moshe Zadka)
Date: Sat, 28 Aug 1999 13:00:42 -0400 (EDT)
Subject: Fuzzy string matching?
In-Reply-To: <y0jso53udoo.fsf@vier.idi.ntnu.no>
Message-ID: <Pine.LNX.4.10.9908281253390.9521-100000@starship.python.net>

On 28 Aug 1999, Magnus L. Hetland wrote:

[about "diff" style algorithms]
> A new and improved version:
[snipped code]

> Could perhaps be a bit more readable, though... (Maybe I'll work on
> that...)

No need, my good man. We have a saying in Hebrew "The work of the rightous
gets done by others"

Added a comment, made a few variables more meaningful, precalculated
ranges, changed some ; to , and here is the changed, and a bit tested,
version:

def distance(a,b):
        n, m = map(len, (a,b))
        if n>m:  # Make sure n<=m
                a,b,n,m = b,a,m,n
        range_m, range_n= map(range, (m, n))
        curr = range(0, n+1)
        for i in range_m:
                prev, curr = curr, [i+1]
                for j in range_n:
                        add, dele = prev[j+1]+1, curr[j]+1
                        change = prev[j] + (a[j]!=b[i] and 1 or 0)
                        curr.append(min(add, dele, change))
        return curr[n]

Well, in the hope I haven't gotten any bugs on the way in, use it to your
hearts satisfaction folks




From dfan at harmonixmusic.com  Thu Aug 19 11:26:59 1999
From: dfan at harmonixmusic.com (Dan Schmidt)
Date: 19 Aug 1999 11:26:59 -0400
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
Message-ID: <wkpv0jkczg.fsf@turangalila.harmonixmusic.com>

Tom Christiansen <tchrist at mox.perl.com> writes:

| Some have mentioned that I've left things out that are in more
| recent releases.  I'm sorry, but I only have Mark Lutz's book to go
| on.  There are no manpages for Python.  Like most people, I'll wait
| until the next edition of the book is release rather than digging
| around elsewhere if the manpage are missing.

I realize they're not manpages per se, but if you haven't already taken a
look at the online documentation at <http://www.python.org/doc/current/>,
you should check it out.  It has a lot of manpage-type information,
although in a different format.  This documentation is also part of
the standard Python distribution.

(no courtesy copies by mail, please)

-- 
                 Dan Schmidt -> dfan at harmonixmusic.com, dfan at alum.mit.edu
Honest Bob & the                http://www2.thecia.net/users/dfan/
Factory-to-Dealer Incentives -> http://www2.thecia.net/users/dfan/hbob/
          Gamelan Galak Tika -> http://web.mit.edu/galak-tika/www/



From jepler at inetnebr.com  Mon Aug 23 08:11:08 1999
From: jepler at inetnebr.com (Jeff Epler)
Date: Mon, 23 Aug 1999 12:11:08 GMT
Subject: assembler optimized integer overflow detection
References: <7B685231630CD311A9B20090276CE41F0B909D@eboxserver2.harc.de>
Message-ID: <slrn7s2er6.4so.jepler@craie.inetnebr.com>

On Mon, 23 Aug 1999 11:28:06 +0200, Felix von Delius
 <Delius at ebox.de> wrote:
>Hi there,
>
>I wondered how Python detects integer overflows and searched the
>corresponding code snippets in Python's source (Objects/intobject.c).
>Because ANSI-C can't detect int overflow errors, there are very strange
>workarounds for all the basic arithmetic.
>
>I asked myself if it will be a big performance boost when someone will
>re-code this routines in assembler, because an overflow can be detected very
>simply by checking a processor-flag after performing the operation. Has
>someone done this already? What speedup-factor would an optimization of this
>kind have to the average Python code? Is it worth the work? (Let's say for
>Linux/i386 machines). I'm not an Intel-assembly guy (only M68K experience),
>but I think this wouldn't be too much efford for an Intel hacker, is it?

How big does this code show up in profiling?  How poor is the produced code
on your target platform?

My guess is that this isn't a big item in the profile, but I could be
wrong.  If it's only, say, 5% and you speed it up by a factor of 5, then
you would gain an overall 4%, not too much.  And I don't know that it is
even 5% on the programs I tend to run.

So in the absence of knowledge that this is a significant consumer of CPU
time, and in the absence of knowledge that there's a way to write the C
better, I'd say there's probably no reason to introduce non-portable
assembly code into Python.

Jeff
-- 
\/ http://incolor.inetnebr.com/jepler/           Jeff Epler jepler at inetnebr.com
Better to light one candle than to curse the darkness.
		-- motto of the Christopher Society



From glen at ranma.interq.or.jp  Tue Aug 17 21:32:24 1999
From: glen at ranma.interq.or.jp (Glen Malley)
Date: 18 Aug 1999 10:32:24 +0900
Subject: Class Description
Message-ID: <37ba0d28@news.interq.or.jp>

Obviously I'm a newbie at this stuff, so go easy on me.
I'm just curious, having not read a great deal of the
object-oriented gear in Python...
Can a class print out info about itself?
For example, I know classes can be dynamically added to...
...could a class re-write itself, if properly messed with?




From mithy at my-deja.com  Thu Aug  5 19:29:45 1999
From: mithy at my-deja.com (Preston Landers)
Date: Thu, 05 Aug 1999 23:29:45 GMT
Subject: downloading web based email
References: <7och4d$jr0$1@nnrp1.deja.com> <14249.55454.471036.787883@dolphin.mojam.com>
Message-ID: <7od6ou$4m1$1@nnrp1.deja.com>

In article <14249.55454.471036.787883 at dolphin.mojam.com>,
  skip at mojam.com (Skip Montanaro) wrote:
>
>     Preston> I'm thinking of writing a Python program to download
email from
>     Preston> web-based services such as Yahoo mail, Deja, Hotmail, and
so
>     Preston> on, and deliever the mail in a format that standard unix
>     Preston> mailers can use.

> Presuming the services you're interested in support a standard mail
transfer
> protocol like POP or IMAP, you need look no further than Eric
Raymond's
> excellent fetchmail program:

I use fetchmail on a daily basis and I'd be overjoyed if it supported
these services.  It doesn't, because they don't use standard IMAP or
POP services.  They use only their HTML interface which is different for
each service.  They want you to look at their banner ads.  Have you ever
used a service like this?  Believe me, I'd use fetchmail if I could.

I've never looked at fetchmail's source, but I believe it to be in C,
and designed to handle a specific class of RFC compliant standard mail
systems.  It would take a major reengineering to make it work like I
want, and I don't feel like doing it in C.  So, unless anybody has any
other suggestions, I'll probably end up coding something that works
largely like fetchmail but is designed to parse results from HTTP based
email services like Hotmail.  Much like a Python program I've seen that
searches Deja News for you.

> If not, perhaps you can train fetchmail to use whatever transport they
do
> offer.

Thanks, but I think "training" fetchmail would involve a nearly from
scratch re-write (in C no less).  A new program in Python sounds much
more pleasant.

--
||  Preston Landers <mithy at my-deja.com>  ||
||        http://askpreston.com          ||


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From claird at starbase.neosoft.com  Sun Aug 22 08:42:05 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 22 Aug 1999 12:42:05 GMT
Subject: How to sell Python to your boss?
References: <ka8yaf7d84t.fsf@delightful.cs.hut.fi> <wjywvur2qrn.fsf@bradh.ne.mediaone.net> <37BD2F89.6CC2@spider.com>
Message-ID: <659B7DD56CD4A360.51878E54B911E668.D1EDB31DC30BE6DF@lp.airnews.net>

In article <37BD2F89.6CC2 at spider.com>,
Alex Maranda  <amaranda at spider.com> wrote:
>Brad Howes wrote:
>> 
>> Petri Mikael Kuittinen <eye at niksula.hut.fi> writes:
>> 
>> > For several months I have tried to get my colleagues excited to learn
>> > Python, but so far nobody has been interested to learn new languages,
>> > except for some mild interest towards Java (which is IMHO an
>> > overly-hyped language).
>Don't sell it. Just go ahead and do it. If your get an assignment you
>can complete more or less on your own, and you're not stuck into a pile
>of legacy code, just do it. Only tell the boss if he asks. No matter if
			.
			.
			.
Sometimes Python is a good answer when one *is* stuck
with "a pile of legacy code".  That is, Python is quite
adept at "gluing", or "wrapping".  When you find your-
self with an intractable application that needs updating,
one of your first reactions should be, "What would happen
if I could access some of this functionality through a
Python interface?"
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From kernr at mail.ncifcrf.gov  Fri Aug 20 12:32:25 1999
From: kernr at mail.ncifcrf.gov (Robert Kern)
Date: Fri, 20 Aug 1999 16:32:25 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <m3ogg3em4k.fsf@atrus.jesus.cam.ac.uk> <37bc4111@cs.colorado.edu> <m3n1vne72r.fsf@atrus.jesus.cam.ac.uk> <37BD1A8A.D4DCFC7C@strs.co.uk> <Pine.LNX.4.10.9908201036330.21136-100000@atrus.jesus.cam.ac.uk>
Message-ID: <37bd821d.5033496@news.erols.com>

On Fri, 20 Aug 1999 10:39:12 +0100, Michael Hudson <mwh21 at cam.ac.uk>
wrote:

>On Fri, 20 Aug 1999, Ian Clarke wrote:
>
>> 
>> > Ah, have you discovered rlcompleter yet? Put this in your ~/.pythonrc:
>> > 
>> > import rlcompleter
>> > readline.parse_and_bind('tab: complete')
>> > 
>> > set PYTHONSTARTUP=~/.pythonrc and then in python type:
>> > 
>> > d={}
>> 
>> Is there any way to get this working on Windoze?
>
>Not that I know of; there is a port of readline to Windows (or at least I
>found a reference to one last time I looked) but I don't know how much
>work would be involved in getting Python to play nice with it.

<plug>
http://starship.python.net/crew/kernr/binaries/Binaries.html
http://starship.python.net/crew/kernr/binaries/PyReadline-w32.zip

It works near-perfectly for me.
</plug>

[snip]

>HTH,
>Michael

Robert Kern           |
----------------------|"In the fields of Hell where the grass grows high
This space            | Are the graves of dreams allowed to die."
intentionally         |           - Richard Harter
left blank.           |



From grahamclemo at my-deja.com  Wed Aug 25 07:18:19 1999
From: grahamclemo at my-deja.com (grahamclemo at my-deja.com)
Date: Wed, 25 Aug 1999 11:18:19 GMT
Subject: Necessity of ``pass''
References: <hhemgu41z7.fsf@forcix.roof.lan> <slrn7s408d.gvn.wtanksle@dolphin.openprojects.net> <hhvha5tswc.fsf@forcix.roof.lan> <14274.54686.281949.239343@dolphin.mojam.com>
Message-ID: <7q0jdl$k53$1@nnrp1.deja.com>

> A similar problem occurs in try/except statements.  If you want
temporarily
> disable the try/except so you can see what exception is being raised,
you
> have to monkey around with the code a bit.  I've gotten used to using
"if
> 0:" and indenting the code another step with C-c > (I use Emacs):
>
>     if foo:
> 	if 0:
> 	    bar
> 	    baz
>     qux
>
> The try/except issue is a little messier.  You wind up with something
like:
>
>     if 1:
>     #try:
> 	try body
>     else:
>     #except:
> 	except body
>

After a bit of experimentation I found you can replace except: with
except 0: (I doubt anyone has raise 0 in their code) and except
ZeroDivisionError: with except 0:#ZeroDivisionError: etc.
(I'm surprised Python silently accepts *any* value in except statements)

> No language is perfect.  I doubt that eliminating the requirement at
least
> one statement in a block is going to help much.
>
> There are probably syntactic considerations as well.  If you allow
empty
> blocks you probably introduce ambiguities in the language that you'd
have to
> correct somehow.

Allowing empty blocks might cause indentation bugs (confusions about
tab sizes for example) to be silently accepted by the parser.

Besides, any expression that has no effect works, such as 0 or '' or
(), albeit less efficiently (it creates extra bytecodes), if you
*really* can't be bothered to type pass.

There's-more-than-one-way-to-do-nothing'ly y'rs,
Graham Clemo


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From claird at starbase.neosoft.com  Wed Aug 25 23:21:16 1999
From: claird at starbase.neosoft.com (Cameron Laird)
Date: 26 Aug 1999 03:21:16 GMT
Subject: Python Performance
References: <3794C642.AB6A3E5D@freenet.co.uk> <p51zdvn4c7.fsf@bidra241.bbn.hp.com> <14236.32616.980578.536614@dolphin.mojam.com> <al8aesin1jr.fsf@sirppi.helsinki.fi>
Message-ID: <FD7584379939819F.69ED398D7189A6FD.5A3332EB837A1A7B@lp.airnews.net>

In article <al8aesin1jr.fsf at sirppi.helsinki.fi>,
Markus Stenberg  <mstenber at cc.Helsinki.FI> wrote:
			.
			.
			.
>Scripting languages are offered as be-all end-all by many people, for some
>reason. 
			.
			.
			.
The usual reasons:  adoption by people who only have
booleans available for their internal calculations,
cagy overcompensation by those trying to balance the
anti-scripters, adolescent exuberance, senile surli-
ness, ...
-- 

Cameron Laird           http://starbase.neosoft.com/~claird/home.html
claird at NeoSoft.com      +1 281 996 8546 FAX



From moritz at news.uni-mannheim.de  Tue Aug 24 14:24:49 1999
From: moritz at news.uni-mannheim.de (Moritz Moeller-Herrmann)
Date: Tue, 24 Aug 1999 20:24:49 +0200
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu> <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net> <siemguw5ar.fsf@cre.canon.co.uk>
Message-ID: <slrn7s5orh.1ck.moritz@Gertrud.chaos.wg>

On Mon, 23 Aug 1999 21:33:32 GMT, Gareth Rees <garethr at cre.canon.co.uk> wrote:
>xah at weborder.com wrote:
>> The ivy tower class of buttheads insist on symbolism for math,
>> clinging to the fancy that certain ideas are better expressed in
>> certain language.  They created a whole barrier of entry [...]

>> [...] If someone cannot read my code, it is their shortcoming.

>Can you spot the contradiction?

Wow! I think people should always shorten posts.

I think the original post is parody though.
I hope so at least. It is, isn't it? Please tell me it is.

-- 
Moritz Moeller-Herrmann mmh at gmx.net ICQ# 3585990        # Not only
Get my public pgp / gpg key from                        # Open Source(TM)
http://webrum.uni-mannheim.de/jura/moritz/pubkeymoritz  # but also
KDE forever! Use Linux to impress your friends!         # Open Minded!




From bvr at cc.pib.alex-ua.com  Wed Aug 25 02:32:04 1999
From: bvr at cc.pib.alex-ua.com (Vadim Beloborodov)
Date: Wed, 25 Aug 1999 06:32:04 GMT
Subject: PyApache / httpdapy on NT
References: <sefw3.928$3U2.2145@dfw-read.news.verio.net>
Message-ID: <005201beeec3$8c7824f0$1935a8c0@dhcp.pib>

>  I've got a python cgi app that's currently running on Solaris
>  with PyApache for a bit of a  performance boost.
>  It looks as if it may have to be moved over to an NT platform.
>  Before I start barking up the wrong tree, is it possible to compile
>  PyApache or httpdapy for Win32? I will need one or the other as
>  I don't think I can afford the cgi overhead.
>  Caveat: VC++ is fairly foreign territory to me, so be gentle.

Yes, is it possible. I am running Apache 1.3.6 and httpdapy on 
Windows NT 4.0.

-- 
Vadim Beloborodov





From bjorn at roguewave.com  Mon Aug 30 09:54:47 1999
From: bjorn at roguewave.com (Bjorn Pettersen)
Date: Mon, 30 Aug 1999 07:54:47 -0600
Subject: The (un)suitability of Make for Python projects (was: Re: Ret
	urning none)
Message-ID: <51C9F3C392D0D111A15600A0C99E64A9AF54E1@localhost>

>     Paul> I know all about make, but in an all-Python project 
> (the best
>     Paul> kind) it is not much value.
> 
> I respectfully disagree.  There is much more to most 
> successful projects
> than building executables.  Building end-user documentation, 
> running tests,
> generating TAGS files, etc. are all appropriate in an 
> all-Python project.

And for some reason I always seem to end up with Python programs writing
Python programs...

who-needs-a-preprocessor-when-you-have-python'ly y'rs -- bjorn



From davidopp at megsinet.net  Wed Aug 25 16:48:30 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Wed, 25 Aug 1999 16:48:30 -0400
Subject: Mirrors?!
Message-ID: <37C4569D.10267F7B@megsinet.net>

Does anyone know where I can download Python 1.52 from somewhere besides
Python.org's site?  I have tried unsucessfully now to download it 3
times...it downloads at .3k per second and fails after around
1600K...arghhh!!!!

Thanks,

David O




From wware-nospam at world.std.com  Thu Aug 19 10:36:44 1999
From: wware-nospam at world.std.com (Will Ware)
Date: Thu, 19 Aug 1999 14:36:44 GMT
Subject: The stupidest question ever seen in this newsgroup...
References: <37BB8F9A.6C204793@rocler.qc.ca>
Message-ID: <FGpv98.CsG@world.std.com>

Gabriel Dupuis (gdupuis at rocler.qc.ca) wrote:
: #!/usr/bin/python
: x = y / 2
: ...other stuff...
: [...NameError...]

Python is complaining that it hasn't heard of a variable called
"y", so when you ask for "y / 2", it can't find it. You could
simply define y like this, _before_ the "x = y / 2" statement:

y = 111298372

But if you do that, you'll need to edit the script every time you
want to factor a new number. So you might want to do something
like this, where you can get the number from the command line:

import sys
y = int(eval(sys.argv[1]))

-- 
 - - - - - - - - - - - - - - - - - - - - - - - -
Resistance is futile. Capacitance is efficacious.
Will Ware	email:    wware @ world.std.com



From ebohlman at netcom.com  Thu Aug 19 17:12:56 1999
From: ebohlman at netcom.com (Eric Bohlman)
Date: 19 Aug 1999 21:12:56 GMT
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <7ph3p8$7ci@junior.apk.net>
Message-ID: <7phs0o$17q@dfw-ixnews16.ix.netcom.com>

Mark W. Schumann (catfood at apk.net) wrote:
: I've seen
: 
: #define ARBEGIN {
: #define AREND   }
: 
: in Clipper (a somewhat C-ish language in this regard) code.
: 
: Eh?

In Clipper, curly braces are used as anonymous array constructors, so 
those mnemonics are presumably short for "array begin" and "array end."




From jon at rdt.monash.edu.au  Tue Aug 31 01:47:42 1999
From: jon at rdt.monash.edu.au (Jonathan Giddy)
Date: 31 Aug 1999 05:47:42 GMT
Subject: Problem with sources - socketmodule.c
References: <37c643ca.348886171@news.u-psud.fr>
Message-ID: <7qfq9u$9vd$1@towncrier.cc.monash.edu.au>

pointal at lure.u-psud.fr (Laurent POINTAL) writes:

>Looking after the error, it was a problem of the number of arguments
>for gethost_by_name_r (and for gethotstbyaddr_r). 
>The ./configure has correctly tested my headers, found a 5 arguments
>gethost_by_name_r, and defined HAVE_GETHOSTBYNAME_R5_ARG.

>But in the socketmodule.c, the source begin by undefining the
>preprocessor symbols HAVE_GETHOSTBYNAME_* set by ./configure, and set
>them depending on the platform... and for linux it automatically set
>it to HAVE_GETHOSTBYNAME_R6_ARG. Which is false on the Slackware4
>Linux.

>Currently, I have put a #if !defined(linux) / #endif around the
>socketmodule.c code which modify HAVE_GETHOSTBYNAME_* and all works
>well.

>Note:
>Linux 2.2.6
>libc.so.5.4.46
>egcs 2.91.66
>gcc 2.7.2.3
>netdb.h v1.4 1995/08/14

The configure tests are not used because they don't work with non-ANSI 
compilers, which are still the default compilers on some platforms.

This sounds like a libc5 vs. glibc problem, but I can't find any details.
If it is, the solution is to add further tests involving the preprocessor 
defines linux and __GLIBC__, as below.  But maybe someone can confirm my
hypothesis first.

e.g.
#ifdef HAVE_GETHOSTBYNAME_R
#if defined(_AIX) || defined(__osf__)
#define HAVE_GETHOSTBYNAME_R_3_ARG
#elif defined(linux) && defined(__GLIBC__)
#define HAVE_GETHOSTBYNAME_R_6_ARG
#elif defined(__sun__) || defined(__sgi) || defined(linux)
#define HAVE_GETHOSTBYNAME_R_5_ARG
#else
#undef HAVE_GETHOSTBYNAME_R
#endif
#endif




From max at max.mcs.gac.edu  Tue Aug 10 09:17:27 1999
From: max at max.mcs.gac.edu (Max Hailperin)
Date: 10 Aug 1999 08:17:27 -0500
Subject: Lambda Functions (fwd)
References: <xpolakis-2708561956530001@athe530-a072.otenet.gr>
Message-ID: <t0u7ln3u654.fsf@max.mcs.gac.edu>

xpolakis at hol.gr (Antreas P. Hatzipolakis) writes:

> ... Why lambda?  ...

We answer this question in the end-of-chapter notes for chapter 1 of
Concrete Abstractions(*), pp. 20-21.  The authority we cite there is
Barendregt(**).  In short, the story is that a circumflex (or "hat")
accent (^) that should have been over the top of the binding occurance
of the variable moved in front of it, and then by visual resemblence
the circumflex with nothing under it was turned into a greek letter
lambda.

That's the interesting part of the story, at least, from Church's
lambda calculi.  The later, less interesting part happened a couple
decades later when come Lisp (pre-Scheme), the greek letter had to get
spelled out in roman letters for the sake of primitive and
anglo-centric computer I/O devices.

(*)  Max Hailperin, Barbara Kaiser, and Karl Knight.  Concrete
     Abstractions: An Introduction to Computer Science Using Scheme.
     Brooks/Cole Publishing Company, 1999.

(**) Henk Barendregt.  "The impact of the lambda calculus in logic and
     computer science."  The Bulletin of Symbolic Logic, 3(2):181-215,
     June 1997.



From g.gruber at xsoft.co.at  Mon Aug 30 03:13:29 1999
From: g.gruber at xsoft.co.at (Gerhard W. Gruber)
Date: Mon, 30 Aug 1999 09:13:29 +0200
Subject: Why do we call python a scripting language?
References: <37c6bb11.340373891@midpoint> <37c66347.518331@news.netset.com> <7q6imd$kuo$1@nntp6.u.washington.edu> <RRiy3.28518$eW1.5110@nntp0.chicago.il.ameritech.net>
Message-ID: <37CA2F19.7357EFDE@xsoft.co.at>

Alexei Boukirev wrote:

> One more explanation of 'scripting' language could be that any program
> written in it requires some environment to run in which is not simply a
> language interpreter but more like a complete usable application itself.

My definition for "scripting" is rather simple. A script is a
substituion for a commandline. When I write a unix shell script I can
execute the same commands in the same order on the commandline and it
will do the same. Python can do the same, but Python is usually not used
in that way. It rather is used (I guess) more like a programming
language. That it is, you write code, test it and if it is satisfying
you use it. Scripting is more similar to working directly on the
commandline to me. Therefore in my opinion python is a programming
language and not a script.
-- 
bye,
   Gerhard
email: sparhawk at eunet.at
       g.gruber at xsoft.co.at



From kistler at gmx.net  Mon Aug 30 18:02:54 1999
From: kistler at gmx.net (Per Kistler)
Date: Tue, 31 Aug 1999 00:02:54 +0200
Subject: Grid (possible scrollable) in Tkinter?
References: <19990830214514.12034.qmail@www0j.netaddress.usa.net>
Message-ID: <37CAFF8E.98FE2C54@gmx.net>

> I'm looking for something like a grid in Tkinter... actually I do not want to
> borrow one from existing libraries, but build one myself from Tkinter pieces,
> if possible. Is this a reasonable thought? :) Would I be able to make the grid
> scrollable (easily)?

Gridding is already there. If a grid complex should scroll, one
might put it into a window embeded in a canvas.

I agree that for oneself, one needs to build a lot of convienent
classes, which do, what one for oneself expects them to do,
like all the scroll variations and toplevel things, but also
for non graphic things.

-Per.
-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From jstevens at basho.fc.hp.com  Fri Aug 13 23:05:30 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:05:30 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132329.RAA07446@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 5:29 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4ceea@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :> They look very different.  It's obvious to the reader what's going on.
> :> They can look at %h and @a and know these are different.
> :
> :Yes.  A flaw, for OO.  
> 
> BFD.

Yup, a big deal.

John S.




From savageb at pacbell.net  Sun Aug 22 18:44:53 1999
From: savageb at pacbell.net (savageb)
Date: Sun, 22 Aug 1999 15:44:53 -0700
Subject: Date Calculations
References: <37BF5D1A.9865CDF@fnmail.com>
Message-ID: <H3%v3.4528$36.461485@typhoon-sf.snfc21.pbi.net>

> I'm making some Date Calculaton class with things like
> deltaDays, addDays, daysInYear and so on, above the calendar
> and time classes. After some 300 lines I thought, I better
> ask, whether that has already been done by someone else.

Yeah, look at:
http://starship.skyport.net/~lemburg/
for mxDateTime. Avery nice package, other interesting stuff there too.

Bob



From mlh at idt.ntnu.no  Fri Aug 27 09:51:03 1999
From: mlh at idt.ntnu.no (Magnus L. Hetland)
Date: 27 Aug 1999 15:51:03 +0200
Subject: Fuzzy string matching
Message-ID: <y0jyaextjqw.fsf@vier.idi.ntnu.no>

Explanation of the distance algorithm...

(Got a request for this, and thought I might as well post it...)

The algorithm:

def distance(a,b):
    c = {}
    n = len(a); m = len(b)

    for i in range(0,n+1):
        c[i,0] = i
    for j in range(0,m+1):
        c[0,j] = j
        
    for i in range(1,n+1):
        for j in range(1,m+1):
            x = c[i-1,j]+1
            y = c[i,j-1]+1
            if a[i-1] == b[j-1]:
                z = c[i-1,j-1]
            else:
                z = c[i-1,j-1]+1
            c[i,j] = min(x,y,z)
    return c[n,m]

It calculates the following: Given two strings, a and b, and three
operations, adding, subtracting and exchanging single characters, what
is the minimal number of steps needed to translate a into b?

The method is based on the following idea:

  We want to find the distance between a[:x] and b[:y]. To do this, we
  first calculate

  1) the distance between a[:x-1] and b[:y], adding the cost of a
     subtract-operation, used to get from a[:x] to a[:z-1];

  2) the distance between a[:x] and b[:y-1], adding the cost of an
     addition-operation, used to get from b[:y-1] to b[:y];

  3) the distance between a[:x-1] and b[:y-1], adding the cost of a
     *possible* exchange of the letter b[y] (with a[x]).

The cost of the subtraction and addition operations are 1, while the
exchange operation has a cost of 1 if a[x] and b[y] are different, and
0 otherwise.

After calculating these costs, we choose the least one of them (since
we want to use the best solution.)

Instead of doing this recursively (i.e. calculating ourselves "back"
from the final value), we build a cost-matrix c containing the optimal
costs, so we can reuse them when calculating the later values. The
costs c[i,0] (from string of length n to empty string) are all i, and
correspondingly all c[0,j] (from empty string to string of length j)
are j.

Finally, the cost of translating between the full strings a and b
(c[n,m]) is returned.

I guess that ought to cover it...

--

  Magnus              Making no sound / Yet smouldering with passion
  Lie          The firefly is still sadder / Than the moaning insect
  Hetland                                       : Minamoto Shigeyuki



From kas at maps.magnetic-ink.dk  Fri Aug  6 12:15:20 1999
From: kas at maps.magnetic-ink.dk (Klaus Alexander Seistrup)
Date: Fri, 6 Aug 1999 18:15:20 +0200
Subject: Help with dictionary, please
References: <1278177468-25600654@hypernet.com>
Message-ID: <slrn7qm2go.3pn.kas@titan.magnetic-ink.dk>

Gordon McMillan wrote:

> > The variable key should be a string, you're supplying a tupple.
> 
> Not at all. Tuples (or any immutable object) make fine dictionary 
> keys.

Sure, I was confusing dictionaries with anydbm objects.  I even tried to
cancel my article but it seems that you were too fast. ;-)  Sorry about
the misinformation.

> The original problem is almost undoubtedly indentation (tabs & 
> spaces, with editor interpretation out of sync with Python's).

Yes, I read your answer.


Cheers,

  //Klaus

-- 
???[ Magnetic Ink ]????????????????????????????????????????????????? ><> ???
???[ Key fingerprint: 2E37 EAED F848 815B 050A FBAA 8F68 DD66 B4B3 7EC4 ]???



From thstr at serop.abb.se  Wed Aug 11 04:13:07 1999
From: thstr at serop.abb.se (Thomas S. Strinnhed)
Date: Wed, 11 Aug 1999 10:13:07 +0200
Subject: any good python source codes i could read?
References: <7op6n6$7qu$1@nnrp1.deja.com>
Message-ID: <37B13093.7DBBFC2B@serop.abb.se>

Hi

For starters: the Python distribution contains quite a lot of it
and not to mention all the modules at:
www.python.org/download/Contributed.html
and maybe Hans Nowak's Snippets-page:
http://www.hvision.nl/~ivnowa/snippets/


Good Luck
 -- Thomas

conv1ct at my-deja.com wrote:
> 
> hi
> i am looking for easy python source codes, or small programs made in
> python
> 
> convict
> 
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.



From 71351.356 at compuserve.com  Mon Aug  2 21:02:49 1999
From: 71351.356 at compuserve.com (T. C. Mits)
Date: Mon, 2 Aug 1999 21:02:49 -0400
Subject: Import a lot of modules, or, import 'foo'
References: <874silabe7.fsf@linuxcare.com> <1%ro3.4357$VL2.467111@news.direcpc.com> <871zdpa9lb.fsf@linuxcare.com> <37A27A54.A737AE62@vision.arc.nasa.gov>
Message-ID: <7o5fb1$lat$1@ssauraac-i-1.production.compuserve.com>

All:
    I see some great replies and extra info given on this NG.   Is someone
moving this to a FAQ or resource or must one do searches all over the place
for it?   If not is deja news the only way to search this NG?

Josef.


Chad Netzer <chad at vision.arc.nasa.gov> wrote in message
news:37A27A54.A737AE62 at vision.arc.nasa.gov...
> "David N. Welton" wrote:
>
> > "Emile van Sebille" <emile at fenx.com> writes:
> >
> > > This seems to work for me:
> > >
> > > >>> for t in tests:
> > >  exec ('import %s' % t)
> >
> > Eewwwww, that's kind of twisted, but I guess if it works:-)
> >
>
> I do a similar thing occasionally, and the big problem is that it makes
> the Traceback system
> work poorly, so that tracking down where unexpected exceptions have
> occurred during
> debugging can be a pain (especially Tkinter or Pmw generated exceptions,
> etc.).
>
> But seriously, you should use the __import__ command instead. ie.:
>
>     string = __import__("string")
>
> which is similar (equivalent at the module level?) to:
>
>     import string
>
> To import names from within the module (ala. "from spam_module import
> eggs"):
>
>     globals()[name] = getattr(__import__(module_name), name)
>
> Chad
>
>





From brad at tera.com  Tue Aug 17 17:15:05 1999
From: brad at tera.com (Bradford Hull)
Date: Tue, 17 Aug 1999 14:15:05 -0700
Subject: CVS Classes in Python
Message-ID: <199908172115.OAA17490@gershwin.tera.com>

If you don't get any answers from Pythoneers, I have some in Perl that 
would be semi-trivial to translate (the languages being similar to such a 
degree) :-)  Of course, it sort of depends on what you want to do; mine 
don't cover everything.
-- 
Bradford K. Hull  | Live as you will wish to have lived when you are dying.
brad at tera.com     |                               Christian F. Gellert
(206)701-2066     | 





From urner at alumni.princeton.edu  Tue Aug 24 15:45:06 1999
From: urner at alumni.princeton.edu (Kirby Urner)
Date: Tue, 24 Aug 1999 19:45:06 GMT
Subject: Pro Python remarks to math ed folks (pointers)
References: <37c2a51e.18050990@news.teleport.com> <Ibjw3.680$v6.61454@typhoon.nyroc.rr.com> <37c2b213.49798667@news.teleport.com> <14274.55328.122770.722670@dolphin.mojam.com>
Message-ID: <37c2f0f7.65901386@news.teleport.com>

Skip Montanaro <skip at mojam.com> wrote:

>Are quadrays similar/identical to quadrics?  If so, why the change in
>nomenclature?
>

No, quadrays and quadrics aren't closely related.  

Quadrays are like the XYZ coordinate system.  XYZ is shaped like 
a "jack" with six spokes to the vertices of a regular octahedron.  

The quadray apparatus, in contrast, consists of four spokes to 
the corners of a regular tetrahedron, labeled (1,0,0,0)(0,1,0,0)
(0,0,1,0) and (0,0,0,1).  The center of the tetrahedron is 
(0,0,0,0) -- this is ordinary volume, not R^4.  

Also I've made the design decision to have this "home base" tetra-
hedron come out with edges of length two (for reasons I maybe 
shouldn't go into here), so Vector(quad=(1,0,0,0)).length() is 
not 1, but root(6)/2

They come closest to being a brand of "simplicial coordinates"
(basing my remarks on a post by Dr. John Conway to geometry_
research) but I'm not sure they fit there either, as we don't
normally normalize to zero, i.e. we don't require a+b+c+d=0
when we write (a,b,c,d) -- just like we don't require this 
of (x,y,z) 3-tuples either.

Here's some dialog in the Python environment to give you 
the flavor:

>>> from vector import Vector
>>> Vector(quad=(1,0,0,0)).length()
1.22474487139
>>> Vector(xyz=(1,0,0)).length()
1.0
>>> v1 = Vector(quad=(1,0,0,0))
>>> v1.data     # default data tuple is in quadray format
(1, 0, 0, 0)
>>> v1.length()
1.22474487139
>>> v1.xyz()     # xyz and polar formats available as conversions
(0.707106781187, 0.707106781187, 0.707106781187)
>>> v1.polar()   # output is (radius,angle,angle), angles in degrees
(1.22474487, 45.0, 35.26438968)
>>> v2=v1*2      # operator overloading (scale vector by 2)
>>> v2.data
(2, 0, 0, 0)
>>> v3=v1+v2
>>> v3.data
(3, 0, 0, 0)
>>> v3.length()
3.67423461417
>>> v4=v3.rotate((0,0,0,120))
>>> v4.data
(0.0, 3.0, 0.0, 0.0)
>>> v4.xyz()
(-2.12132034356, -2.12132034356, 2.12132034356)
>>> v4.polar()
(3.67423461, 225.0, 35.26438968)
>>> v1.angle(v4) # angle between the two vectors v1 and v4
109.47122

For more on quadrays, visit my 
http://www.teleport.com/~pdx4d/quadintro.html -- written in
simple language not presuming a lot of background (target 
audience might be 8th-12th graders).

Kirby

PS:  for those following this thread to the point of trying to
run my code, I should add that I'm importing a couple methods
from NumPy package (free download from Lawrence Livermore Labs).
I know that's a pretty standard add-on, but of course it's not
part of the native Python installation.  Would be easy enough 
to rewrite the bits where I've made use of these methods, but
keeping NumPy in the picture is easier still.

Code is at http://www.inetarena.com/~pdx4d/ocn/python/

Haven't yet written the supporting documentation that would 
make this easier to use.  Mostly just wanting to share my 
enthusiasm, as a curriculum writer, for Python as a teaching
language.  But way cool if people are up for downloading and
testing as well.




From fredrik at pythonware.com  Thu Aug 26 06:27:58 1999
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu, 26 Aug 1999 10:27:58 GMT
Subject: problem compiling extension on Win32
References: <7pvja1$u8e$1@nnrp1.deja.com>
Message-ID: <00b201beefad$cbda88f0$f29b12c2@secret.pythonware.com>

<g_v_wilson at my-deja.com> wrote:
> I am trying to build a small Python extension
> in C on Windows NT using the Microsoft 'cl'
> compiler from the command line (long story).
> Everything is working, except my attempts to
> #define USE_DL_EXPORT and then use DL_EXPORT.
> I have tried #defining the former both before
> and after #including Python.h, but DL_EXPORT's
> expansion is never correct.

iirc, the DL_EXPORT stuff is there to make sure
everything is exported from the python core dll.

if you're building a dynamic extension, and have
trouble exporting the init function, you might do
something like this instead:

void
#ifdef WIN32
__declspec(dllexport)
#endif
initsgmlop()
{
    /* Patch object type */
    FastSGMLParser_Type.ob_type = &PyType_Type;

    Py_InitModule("sgmlop", _functions);
}

(make sure WIN32 is defined when you compile the
module).

also note that you may have to initialize your
type descriptors (if any) in the init function
under Windows; for more info, see:

    http://www.python.org/doc/FAQ.html#3.24

</F>




From tchrist at mox.perl.com  Sat Aug 21 22:00:17 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 21 Aug 1999 19:00:17 -0700
Subject: carpe linguam
References: <199908201554.LAA11430@python.org> <37BDAD0A.FAFC980A@lineone.net>
Message-ID: <37bf4ba1@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Tim Jarman <tim.jarman at lineone.net> writes:
:Of course, having studied Latin, then Spanish, Italian and <WRAP>
:Portuguese are trivial... and partially explains the misquotation <WRAP>
:of Horace in the title of this post.

It was a joke.

But the invective has turned me from this project.  I wanted to learn
Python, and I wanted to get Perl programmers to learn it.  I can see
that the viciousness of assholes (not you) is the same here as elsewhere,
and I'll return now to my previous condition and residence.  Thanks
to the collective you for such an unpleasant experience.    It's not
the language.  It's the assholes.

Bye.

--tom
-- 
#define SVs_RMG   0x00008000  /* has random magical methods */
    --Larry Wall, from sv.h in the v5.0 perl distribution



From clgonsal at keeshah.penguinpowered.com  Sat Aug 28 19:24:10 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Sat, 28 Aug 1999 23:24:10 GMT
Subject: Returning none
References: <000101bef12c$19566520$ad2d153f@tim>
Message-ID: <slrn7sgrse.4ee.clgonsal@keeshah.penguinpowered.com>

On Sat, 28 Aug 1999 04:05:31 -0400, Tim Peters <tim_one at email.msn.com> wrote:
> [C.Laurence Gonsalves]
> > def seqToString( seq ):
> >     ...
> >     result = result + ']'
> >
> > ...
> > Personally, I think falling out of functions and expecting None to be
> > implicitly returned is a very Perl-ish thing to do...
> 
> Oddly enough not.  In the absence of an explicit "return", Perl returns the
> value of the last expression evaluated before falling off the end, so in your
> function above it returns what you intended.

Yes, in this case. I was referring to the way that Perl tries to guess
what you meant rather than treat dubious code as incorrect.

> > Why don't we add $_ to Python as well?
> 
> Because that would be Perlish <wink>.
> ...  
> The reason it doesn't fit in Python is that Python isn't specialized
> to string crunching -- or to any other particular domain.

I was being sarcastic. I was pointing out that if Python is going to
start making wild guesses about one thing, then it might as well do it
everywhere else as well. Making assumptions about what value I want to
manipulate is no less valid than making assumptions about what result I
want to return...

> > I don't think it's too hard to explicitly return None when that's
> > what you want to do, and I would think/hope that most existing
> > Python code does this already.
> 
> Python doesn't return None for the convenience of people too lazy to
> explicitly return None themselves; it's instead a cheap way to make it
> probable that someone using a procedure as a function will get an
> error (None supports no operations except true/false interpretation)
> without blowing up the interpreter (do this in C and you'll access
> random stack trash).

That sort of argument also says that when I go outside the bounds of a
list in Python, I should get back None, since in C I would access random
stack/heap trash. Instead, Python raises an exception. I think it should
do the same if you don't return anything, and then try and access that
nothing.

While None does have few operations, the current behaviour does not
guarantee a runtime error. I've very often built-up sequences filled
with None's, or assigned None to the attributes of some object. The
program ends up failing much later. I think it's best for Python to fail
as close to the source of the problem as possible. Ironically, this
"failing much later" is the same kind of behaviour you get with C
programs that access random stack trash...

That's why I also suggested some other possible rules that would enable
some amount of compile-time checking, so that function writers who do
questionable things would be alerted.

> > It is possible, but more complicated, to make this a compile time
> > error.  This would essentially require a function/procedure
> > distinction with restrictions about how you used return in each one.
> > ...
> 
> Do you imagine Guido didn't consider this when he designed the language?

So Python is *absolutely perfect* and shouldn't be changed at all? I
guess us mere mortals shouldn't even bother making any suggestions then.
Python 2 will just be Python 1.5.2 with a new name...

In another post I suggested another idea that requires little or no
modifications to the grammar, but would just add some new runtime and
compile time checks. Functions using return (or falling off the end)
would return nothing at all (rather than None), and function definitions
would require that the function either always return a value, or never
returns a value. This can be very easily checked at compile time. At
runtime, attempting to retrieve the return value of a function that
returns nothing would raise an exception. This check could even be done
before actually executing the function at all.

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From jozart at go2net.com  Thu Aug  5 21:35:42 1999
From: jozart at go2net.com (Joe Bowbeer)
Date: Thu, 05 Aug 1999 18:35:42 -0700
Subject: load .py file in interpreter?
Message-ID: <37AA3BEE.9417D35A@go2net.com>

Once you're running in the interpreter, how do you load (or re-load)
some foo.py Python code into the interpreter? (I'm using JPython, by the
way.)

This seems like something basic for interactive development. I must be
missing something obvious.

--
Joe Bowbeer





From sjturner at ix.netcom.com  Tue Aug 17 13:15:41 1999
From: sjturner at ix.netcom.com (Stephen J. Turner)
Date: Tue, 17 Aug 1999 13:15:41 -0400
Subject: Getting the module name from a method...
References: <7pboln$n74$1@feed.teaser.fr>
Message-ID: <37B998BD.1A35F74B@ix.netcom.com>

Olivier Deckmyn wrote:
> I would like a method of my own to be able to print the module it is
> declared in:
> 
> ex:
> the text for the module : toto.py is :
> 
> class MyClass:
>     def myMethod(self):
>         print "I am in module", ????????
> 
> and then
> MyClass().myMethod()
> should produce :
> 'toto'

How about "self.__class__.__module__"?

Regards,
Stephen

--
Stephen J. Turner <sjturner at ix.netcom.com>



From bruce_dodson at bigfoot.com  Sat Aug 14 03:19:41 1999
From: bruce_dodson at bigfoot.com (Bruce Dodson)
Date: Sat, 14 Aug 1999 04:19:41 -0300
Subject: Help: sendmail under MS NT
References: <7p1pma$ak6$1@nnrp1.deja.com> <37b4ca58.2777926@nntp.best.ix.netcom.com>
Message-ID: <XM8t3.62641$jl.39581340@newscontent-01.sprint.ca>

Check the topic indexes at C/C++ User's Journal and Dr. Dobb's Journal web
sites.  Sometime between mid 1996 and early 1998, there was an article
featuring a sendmail DLL for Win32, which hid the messy internals of MAPI
and left you with something basically as simple as sendmail.  It would be a
trivial matter to encapsulate this and the Unix sendmail in a portable
least-common-denominator wrapper.




From mryan at netaxs.com  Wed Aug 18 13:38:39 1999
From: mryan at netaxs.com (Michael W. Ryan)
Date: 18 Aug 1999 17:38:39 GMT
Subject: Python Sighting
Message-ID: <slrn7rlrsu.l2e.mryan@unix2.netaxs.com>

I made an interesting discovery the other day while roaming through the
registry of an NT system trying to troubleshoot a problem in a Microsoft
Site Server class I was supporting:  Microsoft Site Server 3.0 (not sure
if it's the normal or commerce edition) uses Python.  From the presence of
the ni.pyc file, I'm going to guess that it's 1.4.x.  I'm also not sure of
what capacity they're using it in.

-- 
Michael W. Ryan, MCP, MCT     | OTAKON 1999
mryan at netaxs.com              | Convention of Otaku Generation
http://www.netaxs.com/~mryan/ | http://www.otakon.com/

PGP fingerprint: 7B E5 75 7F 24 EE 19 35  A5 DF C3 45 27 B5 DB DF
PGP public key available by fingering mryan at unix.netaxs.com (use -l opt)



From fredp at multimania.com.nospam  Thu Aug 26 12:14:53 1999
From: fredp at multimania.com.nospam (Fred Pacquier)
Date: 26 Aug 1999 16:14:53 GMT
Subject: Ann: HTMLized calendar module
Message-ID: <8E2EB09C3PaCmAnRDLM@news.dial.oleane.com>

FWIW:

For a small project I did at work I needed to create web pages with 
calendars that could be modified by a python script on a regular basis 
(like adding links to month names, days, week numbers...).

I thought I would find something similar ready-made on the net, but I 
didn't. So I took the standard library module calendar.py and turned into 
html_cal.py.

There is only one new method that creates a monthly calendar like 
prmonth(), except that:

- it prints week numbers
- it returns a string instead of printing to stdio
- month names, weekdays and week numbers are tagged with "pseudo-HTML", ie 
tags that will not display until some other script has changed them to 
actual <a href=""> or such.

It's probably not the purest python code in the world (mine, I mean :) as 
I'm quite new to this, but it does the job. If it may be of use to anyone, 
just mail me (beware the .nospam :). If it's already been done better, 
please enlighten me, too.

-- 
YAFAP : http://www.multimania.com/fredp/



From gmcm at hypernet.com  Mon Aug  9 10:53:13 1999
From: gmcm at hypernet.com (Gordon McMillan)
Date: Mon, 9 Aug 1999 09:53:13 -0500
Subject: An idea - request for comments
In-Reply-To: <37AED27F.1C3D7CC9@houbits.com>
Message-ID: <1277916098-5086849@hypernet.com>

Nigel Head wrote:
[licensing...]

> I guess what would really bug me is for people to hijack the
> package, remove original credits and resell straight with no added
> value of their own. I think that should be disallowed; that's all!

That's disallowed by copyright law, which is how violators get 
prosecuted anyway, not under these silly "my license is trickier than 
yours" clauses.

- Gordon



From arcege at shore.net  Thu Aug  5 10:27:39 1999
From: arcege at shore.net (Michael P. Reilly)
Date: Thu, 05 Aug 1999 14:27:39 GMT
Subject: "hello, world" of embedding/extending?
References: <7o9sum$2ak$1@news.panix.com>
Message-ID: <vbhq3.412$EG4.57242@news.shore.net>

kj0 <kj0 at mailcity.com> wrote:

: What are the "hello, world" equivalents of embedding and extending
: Python (in/with C or C++)?

: More generally, where can I find source code didactic examples of
: embedding and extending Python?  (I write "didactic" because I'm not
: quite at the level yet of being able to read real-world,
: multiplatform, distribution-grade applications; rather, I'm interested
: in examples ranging from the "hello, world" level of complexity to the
: level of a full-blown useful application that someone may write for
: personal use, i.e. not intended for wide distribution).

The basic equivalents are in the Demo/embed/ and Demo/extend/ directories
of the source distribution.  For full blown extensions just take a look
at any of the C modules in the Modules/ directory.

There is documentation for embedding Python at 
  http://www.python.org/doc/current/ext/embedding.html and
  http://www.python.org/doc/current/api/veryhigh.html
And for extending at
  http://www.python.org/doc/current/ext/ext.html and
  http://www.python.org/doc/current/api/api.html and
  http://starship.python.net/crew/arcege/extwriting/pyext.html

The last webpage (tersely) describes how to make extension types.

  -Arcege




From clgonsal at keeshah.penguinpowered.com  Sun Aug 22 18:09:18 1999
From: clgonsal at keeshah.penguinpowered.com (C.Laurence Gonsalves)
Date: Sun, 22 Aug 1999 22:09:18 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <slrn7s0t7v.o5.clgonsal@keeshah.penguinpowered.com>

On 19 Aug 1999 08:22:49 -0700, Tom Christiansen <tchrist at mox.perl.com> wrote:
> GOTCHA: (medium)
>     Perl's hex() function is the opposite of Python's.  Perl's hex
>     function converts 22 into 37.  Python's hex function converts 37 into
>     0x22.  Oh my!  Like hex, Python has oct() completely backwards from
>     Perl.  Python's oct(44) returns 054, but Perl's oct(44) return 36.

I'd have to say Python definitely has more sensible meanings for these
functions. Consider the following:

In Perl:
    hex(10)        -> 16
    hex(0x10)      -> 22

Okay, so hex(10) is 10 reinterpreted as hex, but what is hex(0x10)? It's
0x10, converted to decimal (16), and the reinterpreted as hex (0x16 =
22). Huh?

In Python:
    hex(10)        -> '0xa'
    hex(0x10)      -> '0x10'
 
In Python, hex converts a number into a string containing the hex
representation. Much clearer than the rather circuitous meaning of the
Perl hex function...

I suppose the Perl hex function's weird behaviour has to do with the
fact that 'scalars are scalars' in Perl, and numerical quantities are
treated as equivalent to strings containing their decimal
representation. So in Perl, 0x10 == 16 == "16", and the hex function is
really intended to convert strings containing hex quantites into
numbers. It's rather odd that in Perl, 0x10 is equivalent to the string
"16", so passing 0x10 to Perl's hex results in decidely non-intuitive
behaviour.

In Python, if the hex function were to do anything at all like Perl's
(and still be at all useful), it would have to take a string as a
parameter, not a number, in which case you'd be complaining that hex(10)
raised a TypeError. It wouldn't make sense for it to take a number,
because how do you convert a number 'from hex'? That has no meaning,
unless you treat numbers as decimal strings (like Perl does).

In Python, the meaning of hex is also conceptually consistent with
type-conversion functions like int, float, str, etc. In Python,
conversion functions convert *to* what the function names, not from it.

-- 
  C. Laurence Gonsalves            "Any sufficiently advanced
  clgonsal at kami.com                 technology is indistinguishable
  http://www.cryogen.com/clgonsal/  from magic." -- Arthur C. Clarke



From m.faassen at vet.uu.nl  Sat Aug 21 21:15:17 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 22 Aug 1999 01:15:17 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <7pmdn5$lt0$1@nnrp1.deja.com>
Message-ID: <7pniv5$mlm$2@newshost.accu.uu.nl>

srenner at lycosmail.com wrote:
> The reactions Mr. Christiansen evokes may have more to do with his style
> than his content. It strikes me as pompous, pretentious, hostile,
> autodidactic, preening, vain and unpleasant. Perhaps this is not the
> right newsgroup for him.

Oh, I don't think he's that bad. He *does* flame back when he feels people
are flaming him, and sometimes his finger is a bit too near the flame
trigger, but he's nice and reasonable to the people who are nice and 
reasonable to him.

So let's try to be nice and reasonable to him? I thought this group was
above this kind of flaming, actually. I always considered comp.lang.python
to be one of the most mature newsgroups I've been reading, but I'm slowly
being forced to adjust my opinion here, and I don't like doing that.

Regards,

Martijn




From tchrist at mox.perl.com  Sat Aug 21 12:13:38 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 21 Aug 1999 09:13:38 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <7ple6r$kf0@ds2.acs.ucalgary.ca>
Message-ID: <37bec222@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    nascheme at ucalgary.ca (Neil Schemenauer) writes:
:It is not too hard to remember what is mutable and what is not.

I'm not sure that this maps well to the way people think, considering
that everyone keeps telling me that it's "just something you have to
learn and/or get used to", which means it's not intuitive at first.
And that was my point of listing it as a gotcha.

:>    You can't use "in" on dicts.  Instead, you must use 
:In the Python philosophy, every expression should have an obvious
:meaning.  "in" in this case does not.  Perl has a different
:philosophy here.

To me it would be clear that it's a wasteful linear search of the
values, just as in a list it's a wasteful linear search of the elements.
The point wasn't a request for a feature, just an observation of a
conterintuitive point in something that would otherwise seem to be a
natural effect.

:The rule for determining if something is local is quite simple.
:If a variable is bound anywhere in a function then it is local
:unless declared global by the "global" statement.  

I understand.  But it's well, surprising that it's all context-dependent.
Not fun to look for.

:>    I feel that this hides the real test condition in a place
:>    it doesn't belong.
:
:On the up side, because assignment is a statement you can bind
:multiple names at once:
:
:	x = y = z = 0.0

That's not a side-effect of this.  You can do that in most languages,
including C and Perl and the like.  

:Again the semantics of such statements is not immediately obvious
:so Python avoids them.  They may be eventually added to the
:language though.

People keep saying this, but I've never understood why 

    a += b

isn't immediately obvious as

    a = a + b

:
:As I mentioned earlier, everything is a reference.  You just have
:to know what is mutable and what is not.  Lists, obviously from
:your example, are.

Alas, I'm pretty sure that this isn't how people think unless
they're trained to do so.  There's a reason that in Perl most
people still use 

    @a = (1,2,3);

not

    $a = [1,2,3];

It's because @a is a first-class array, and $a merely a reference
thereto.  Copying things not producing a copy is well, not what
non-serious programmers immediately think of.  Why can python copy
sequences of chars (1-byte strings) trivially:

    s = "this stuff"
    t = s

But as soon as you have a sequence of, oh, integers for example, you
can't use the same set-up.  Very odd.

:>    Often Python's error messages leave something to be desired.
:>    I don't know whether 
:
:I don't find them too bad.  I guess I don't use Perl much so I
:don't know what I am missing.  At least they are better than
:SEGV. :)

See my longer response to Tim Peters about this elsewhere in the group.
It's not exactly about this, but demos what I mean.

:>GOTCHA: (medium)
:>    All ranges are up to but *not including* that point.  So range(3)
:>    is the list [0,1,2].  This is also true in slices, which is the
:>    real gotcha part.  A slide t[2:5] does not include t[5] in it.
:
:This is explained quite well in the tutorial.  It makes a lot of
:sense once you get used to it.

As before, "once you get used to it" is the key point here.

:I think you are a bit confused by tuples.  In the first example
:you are looping over a tuple of length two.  In the second case
:you are looping over a string.  A length one tuple must end with
:a comma.  This is weird but necessary for the parser to tell the
:difference between a tuple and a parenthesized expression.

You said it yourself: this is weird.

:Again with the references. :)  Don't think references.

You can't avoid it.
:Does Perl really make a copy of a list every time you assign it
:to another variable?  How about other things like instances?

Yes, with 1st-class arrays (you say lists) and hashes (you say
dictoinaries).  But not with instances, which are always by definition
references, and suffer the same confusion.  That's why base types
are easier to use for many "accidental programmers".

:>    Slices in Python must be contiguous ranges of a sequence.
:>    In Perl, there's no such restriction.  
:Use a dictionary if you want this.

Huh?  How would I use a dictionary (don't you get tired of such a long
and non-intuitive word?) to do this:

    @a[3,7,4] = ("red", "blue", "green")
or
    @a[3,7,4] = @a[2,9,10];

:>    You can't slice dictionaries at all in Python.  In Perl, it's
:>    easy to slice a hash, and just as sensible.
:
:The semantics of this are not immediately clear.

Sure they are.  

    @hash{"red","blue","green"} = ("rojo", "azul", "verde")

is the same as 

    ($hash{"red"}, $hash{"blue"}, $hash{"green"}) =
	("rojo", "azul", "verde")

which is the same as:

    $hash{"red"}   = "rojo";
    $hash{"blue"}  = "azul";
    $hash{"green"} = "verde";


I'm only fibbing a little.  Assignments of slices look for overlap, just
like

    ($a, $b) = ($b, $a)

does.

:del is a statement.  It has to be so it can remove variable
:bindings.  

I *really* don't undertsand this. Why does Python have so many
"statements" compared with C or Perl?  What does this actually buy you?
I can remove variable bindings just fine with the undef() or delete()
functions in Perl, without needing to sacrifice an entire *statement*
in the grammar for this!  There must be something important that I'm
not understanding here, and I'd like to.

Oops, time's up.   Off to teach OO Perl at the Monterey conference. :-)

--tom
-- 
You know, by the time you get done with all this, the "Swiss Army
Chainsaw" is going to be more like a Swiss Army Tactical Nuke....  :-)
    --Brandon Allbery on perl in <1991Feb21.002412.20045 at NCoast.ORG> 



From fdrake at acm.org  Thu Aug 12 11:13:27 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Thu, 12 Aug 1999 15:13:27 GMT
Subject: Python GUI
In-Reply-To: <934408001snz@vision25.demon.co.uk>
References: <7ni69d$nc7$1@nnrp1.deja.com>
	<37B0E67D.27F8D2ED@prescod.net>
	<934408001snz@vision25.demon.co.uk>
Message-ID: <14258.58519.458528.761079@weyr.cnri.reston.va.us>

Phil Hunt writes:
 > I think I will have to have a look at Glade. My understanding is that 
 > it uses its declarative file format to save/load UI definitions that
 > people are working on.

  A couple of us here have started playing with it, and it's pretty
neat.  It does create an XML representation of the project as its own
storage format, but an application can use libglade to load the XML
and create widgets on the fly.  The latest release of PyGTK /
gnome-python includes bindings to libglade, and it seems to work
"OK".  There are some issues related to Gnome widgets, but otherwise
seems pretty good.  (I expect the issues will be resolved fairly
quickly.)

 > If Parrot could read/write Glade's format, that would be good, as
 > then people could write a UI in Glade and export it to any
 > backend that Parrot supports.

  That would be neat!

 > That is a persuasive argument. One counterargument would be that 
 > any application that uses that approach would need a built-in
 > XML parser and Parrot run-time library, thus creating code bloat;
 > however, Parrot is explicitly intended to support multiple back-ends,
 > so there is no reason it couldn't support both approaches.

  There's also the issue of not wanting to store the user interface
information in an external file; this is probably more of an issue for 
commercial software than open source materials, but there are valid
reasons for "locking down" the interface.  It could be implemented by
including the XML in the binary as a string or by generating "native"
code to do the right thing.

 > > > True; I don't think a parser for Python will be too difficult to
 > > > write: my provisional spec only has 12 productions.
 > > 
 > > These specs always start out small. :)

  Are you aware of Don Beaudry's "pgen" module?  It provides a runtime
Python interface to the parser generator used for Python, so it's easy 
to create "Pythonesque" languages fairly easily.  If you want an
easy-to-read langauge for Parrot, you might consider that.  (I'm not
sure of it's status with respect to current versions of Python; poke
around ftp.python.org:/pub/python/contrib for the latest version.)


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From tchrist at mox.perl.com  Thu Aug 19 15:13:45 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 19 Aug 1999 12:13:45 -0700
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <8E27B3FC4duncanrcpcouk@news.rmplc.co.uk>
Message-ID: <37bc4959@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    duncan at rcp.co.uk (Duncan Booth) writes:
:However I agree it can be confusing that when you subscript a string you 
:get another object of type string.

*Tell me* about it!

    >>> s = "fred"
    >>> print s[0]
    f
    >>> print s[0][0]
    f
    >>> print s[0][0][0]
    f
    >>> print s[0][0][0][0] 
    f
    >>> print s[0][0][0][0][0]
    f
    >>> print s[0][0][0][0][0][0]
    f

--tom
-- 
"... my dear old friend. I wish to God there were more automata in the world
 like you." - Charles Darwin to T H Huxley, letter, 1882.



From tchrist at mox.perl.com  Sat Aug 14 21:58:41 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 14 Aug 1999 18:58:41 -0700
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk>
Message-ID: <37b610c1@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Robin Becker <robin at jessikat.demon.co.uk> writes:
:Lets face it; the perlers have a fairly hopeless position when it comes
:to defending the perl language 

I'm going to pretend you didn't just now so egregiously underestimate
both the acumen and the tenacity of those whom you don't know, for
any other response would lead inexorably to ratholes so tortuous as to
captivate Borges, so black as to plunge Pollyanna into deepest despair.
For your own personal benefit, however, you might in the future take
some care to avoid repeating that particular stultiloquy. :-)

--tom
-- 
"Espousing the eponymous /cgi-bin/perl.exe?FMH.pl execution model is like 
reading a suicide note -- three days too late."
	    --Tom Christiansen <tchrist at mox.perl.com>



From tim_one at email.msn.com  Tue Aug 24 02:19:32 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 24 Aug 1999 02:19:32 -0400
Subject: "def" vs "sub" (was RE: More random python observations from a perl programmer)
In-Reply-To: <19990823095217.D28043@xs4all.nl>
Message-ID: <000601beedf8$a1ca13c0$902d2399@tim>

[Thomas Wouters]
> ...
> Nevertheless, compiler and language internals fascinate me.

With luck you'll grow out of it <wink>.

> ...
> Which reminds me... I hardly see any 'development-traffic' on this
> list, other than some sideways references to a CVS tree... Is there
> a secret society or do you just have meetings in guido's favorite
> pub ?

There are many secret societies; most development takes place in Guido's head,
while most of the rest occurs under the auspices of one of the Secret Insider
Groups (SIG, for short):

    http://www.python.org/sigs/

Python's source code is exceptionally easy to understand, so pick something
that interests you from

    http://www.python.org/cgi-bin/todo.py

and have at it.

we'll-beat-the-curiosity-out-of-you-yet<wink>-ly y'rs  - tim





From m.faassen at vet.uu.nl  Sat Aug 21 21:22:38 1999
From: m.faassen at vet.uu.nl (Martijn Faassen)
Date: 22 Aug 1999 01:22:38 GMT
Subject: carpe linguam
References: <37bd5e2a@cs.colorado.edu> <rnAv3.3533$qt5.98359@iad-read.news.verio.net>
Message-ID: <7pnjcu$mlm$3@newshost.accu.uu.nl>

Tom Culliton <culliton at clark.net> wrote:
> In article <37bd5e2a at cs.colorado.edu>,
> Tom Christiansen  <tchrist at mox.perl.com> wrote:
>>Private mail has indicated a remarkably amount of distrust about my
>>learning Python.  Some verges on the unbelievable.  

> Tone.  Everything has to do with tone.  If you read a comparison of
> Perl to some other language written in the same tone as your original
> Python/Perl messages you might bristle too.  Your tone makes it sound
> like you're trashing Python, and often arbitrarily, "this isn't done
> in the same way thet perl does it so it's gross..." often seems to be
> the sum and total of your objection in the original message.  You do
> tone it down some in the next one, and even point out things that you
> think are "cool" about Python, but that write up still often misses
> the point because of gaps in your understanding.

But even if Tom's tone is sharp, which I don't deny, why can't we just
deal with that? And *of course* Tom's coming from Perl, and he is looking
at Python with Perl eyes. This is to be expected. So point out to him
where you think his misconceptions are.

>>Secondly, and much more importantly, the reason I am learning Python
>>is *not* because I wish to "score some quick hits and prove to c.l.p.m
>>that Perl is best after all" [sic] as one private querent asked, to my
>>complete and utter horror.
>>
>>You may not have known this of me, but I am a collector of languages,
>>and to some extent, of language trivia.  Can you honestly believe that
>>I studied Spanish, French, German, Italian, Portuguese, or Latin merely
>>to prove to myself and the world that English was the best language?

> I would assume your motivations for learning languages are the same as
> mine, to broaden your horizons, pick up useful ideas and concepts, be
> able to understand what other folks are on about, ...  All noble and
> admirable goals.

> Your motivation, as a prominent Perl advocate, for publishing these
> comparisons, while you're still coming up the learning curve, seems
> more questionable, and appearance is usually all that we have to go by
> on the net.

So on the net we take the best interpretation possible, which is that Tom's
trying to share his learning experience with us. This way we can learn
more about Perl (and Tom's way of thinking, incidentally), and *Python*, and
we can help him where we think he's wrong.

> Every language has it's oddities, rough spots, and abuses.  Every
> language has it's own idioms, tricks, and required knowledge.  Every
> language has it's advocates and detractors.  Some of your critiques of
> Python are valid, some are just incomplete knowledge, some are just
> matters of taste or what you're used to.  Since you are who you are,
> your comparison got widely published by LWN and possibly others as
> "gospel" which is rather distressing.

A while ago I was struggling to learn Perl. If I'd found those comparisons
at the time I'd certainly take a look at Python. I did in any event and 
liked it. People can form their own opinions. This is the net, we're hackers,
we have opinions. :)

> Frankly I generally skip anything in this newsgroup with Perl in the
> title as a likely flame fest.  I only saw your messages because LWN
> had published them.  Since most of what I've seen you post in the past
> seems rational, reasonable, and intelligent, my take has been that
> you're somebody who has interesting and worthwhile things to say, if
> with a general Perl bias.  So maybe you were just having an off day
> and weren't minding your style?  ;-) ;-) ;-)

Oh, much better interpretation of things already. :)

> I urge you to keep trying and keep writing about your experience with
> learning Python.  It's good for the Python community to get different
> perspectives on the language, and discuss them.

Oh, why did you moderate this latter end of your post so much? My
whole reply down the drain as you say the same yourself! :)

> I also urge you to
> double check the details, and avoid potential flamebait when you do
> so.  If we can't get by your style we can't appreciate your substance.

I feel I can live with his stuff now already. Substance over style. :)

Regards,

Martijn




From zehp at francemel.com  Mon Aug 23 07:20:48 1999
From: zehp at francemel.com (Mathieu Lecarme)
Date: Mon, 23 Aug 1999 13:20:48 +0200
Subject: Authentication with httplib
References: <7pr09c$dag$1@news.netvision.net.il>
Message-ID: <7prana$a0p$1@minus.oleane.net>

> I am using httplib to access a Page on a Server:
>
>     .....
>      h = httplib.HTTP(Server)
>      h.putrequest('GET', Page)
>      h.endheaders()
>     .....
>
> When Server is configured not to require a passwd, I set
Server="myServer",
> Page="/myPage",
> and everything is OK.
>
> My Q regards the situation where there is a need for authentication: Where
> should I put the Username+Passwd?

you should rather use urllib :
    import urllib
    file = urllib.urlopen('http://login:password at www.foo.com/truc.html')

or if you wont to know how it's work, have a look of urllib line 248 and
after

Mathieu





From roy at popmail.med.nyu.edu  Sun Aug 22 11:00:06 1999
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Sun, 22 Aug 1999 11:00:06 -0400
Subject: More random python observations from a perl programmer
References: <000001beec6d$9753e240$542d2399@tim> <m2iu677ujv.fsf@netmeg.net>
Message-ID: <roy-2208991100070001@mc-as01-p27.med.nyu.edu>

Les Schaffer <godzilla at netmeg.net> wrote:
> what does orthogonality mean here?

Two concepts/variables/things/whatever are orthogonal if either can assume
any legal value with no effect on the other.

For example, color and style, which describe a house.  Color can be any of
red, blue, green, brown, or white, and style can be any of colonial,
ranch, victorian, or cape_cod.  Any of those styles of house can be
painted any color, so color and style are orthogonal.

But, let's say we had two other variables describing the house,
roof_material, and roof_color.  You can have the same assortment of colors
for the roof as above, and the materials can be shingle, ceramic_tile, or
metal.  Assume that ceramic tiles only come in red, green, or brown. 
Roof_material and roof_color are NOT orthogonal; while color can be white
or green, it can only be so if material is not ceramic_tile.

So it is with types and operators.  Let's say a language has three types,
int, float, and string, and has a bunch of operators, add, subtract,
multiply, etc.  The language is orthogonal if any operator can apply to
any type(s).  Some operators in python could apply to ints, floats, and
strings, so any of:

x = 4 + 3
x = 4.0 + 3.0
x = "Four" + "Three"

are legal, with x getting the value 4, 4.0, and "FourThree" respectively. 
But, not all operators can be applied to all types.  Thus:

x = 3 * 4

is legal, as is:

x = "Three" * 4

but:

x = "Three" * "Four"

is not.



From scarblac-spamtrap at pino.selwerd.cx  Sun Aug 22 19:52:23 1999
From: scarblac-spamtrap at pino.selwerd.cx (Remco Gerlich)
Date: 22 Aug 1999 23:52:23 GMT
Subject: What is a troll by the way? [offtopic]
References: <Z1xv3.11$EN1.7779@news.shore.net> <99082115204502.12287@o112>
Message-ID: <slrn7s137e.4s9.scarblac-spamtrap@flits104-37.flits.rug.nl>

Fredrik Henbjork <frehe491 at student.liu.se> wrote:
> WH0 KANN F0R63T ZUCH AN AWZUM K001 D00D!!!!! BIFF R00LZ!!!!

ITYM "!!!!!!!1111". HTH. HAND. ;-)

And it's B1FF, iirc...

Too bad that's all before my time, I only know the stories
from the Jargon File.
-- 
Remco Gerlich,  scarblac at pino.selwerd.cx



From paul at prescod.net  Fri Aug 20 11:21:25 1999
From: paul at prescod.net (Paul Prescod)
Date: Fri, 20 Aug 1999 15:21:25 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <37BD7275.FE2B56EF@prescod.net>

Tom Christiansen wrote:
> 
> ...
>
> GOTCHA: (low)
>     When you need to make an empty copy of the same type, you write
>         x = y[:0]
>     So much for being typeless. Sigh.

I don't think that Python is a typeless language. There is no such thing
as a typeless language: some languages have a single type but they have
at least one type.

Python is a dynamically typed language, just as Perl is. In some cases
it is less "in your face" about types, but it is still a typed language.

 Paul Prescod



From greg.ewing at compaq.com  Tue Aug 17 17:12:12 1999
From: greg.ewing at compaq.com (Greg Ewing)
Date: Wed, 18 Aug 1999 09:12:12 +1200
Subject: Regarding the magical "or" operator
References: <37b5d712@cs.colorado.edu>
Message-ID: <37B9D02C.A034B044@compaq.com>

Tom Christiansen wrote:
> 
> My question is this: have you ever had an urge for it (or a variant)
> to care about the difference between None and any other false value?

Personally I think there should be a separate Boolean type,
and it should be an error to test the truth of any other type.

Alternatively, the Scheme convention could be used, where
the one and only false value is the boolean one, and everything
else is true.

There should be an "else" operator for doing what the "or"
trick is currently used for. The expression

   a else b

should return a unless it is None (and only None), in which
case it should evaluate and return b.

(I don't like the idea of calling it "nor", because that
word already has a meaning in computing which is quite
different.)

Greg



From mryan at netaxs.com  Mon Aug  2 14:00:02 1999
From: mryan at netaxs.com (Michael W. Ryan)
Date: 2 Aug 1999 18:00:02 GMT
Subject: CGI Database Question
References: <37a33a28.23844403@news.ne.mediaone.net>
Message-ID: <slrn7qbn51.9u.mryan@unix2.netaxs.com>

On Sat, 31 Jul 1999 18:05:20 GMT, Joel Gould <joelg at alum.mit.edu> wrote:

>I would like to use Python for CGI scripts.  I am going to ask my ISP
>to install Python (they run an NT server with IIS), but what database
>package should I use?  Specifically, what should I tell my ISP to
>install so I can use a (simple) database from my CGI scripts?

The Win32 distribution (I believe) comes with a dbm implementation.  The
dbm family of flat file databases are good for lightweight file-server
databases, and are the database type used by shelve.

>Also, do I have to worry about thread safety?  In other words, can
>multiple requests come into the web server causing multiply copies of
>my Python scripts to be run concurrently?  Does this impact the
>database selection?

I'm not sure how dbms handle file locking on the database files, but I
think you have to handle it on your own.  Generally, CGI invokes a
separate Python interpreter process for each CGI request.

-- 
Michael W. Ryan, MCP, MCT     | OTAKON 1999
mryan at netaxs.com              | Convention of Otaku Generation
http://www.netaxs.com/~mryan/ | http://www.otakon.com/

PGP fingerprint: 7B E5 75 7F 24 EE 19 35  A5 DF C3 45 27 B5 DB DF
PGP public key available by fingering mryan at unix.netaxs.com (use -l opt)



From gweon at umich.edu  Thu Aug 12 11:14:46 1999
From: gweon at umich.edu (Gey-Hong Gweon)
Date: 12 Aug 1999 11:14:46 -0400
Subject: converting Netscrape mailbox to Unix mbox format
References: <14256.35316.328575.118034@lrz.de> <m3wvv24wcg.fsf@atrus.jesus.cam.ac.uk> <14257.45321.128760.603105@lrz.de> <37B1BD55.FA34A1CC@callware.com>
Message-ID: <m2btcdkp3t.fsf@allenguest.physics.lsa.umich.edu>

Ivan Van Laningham <ivanlan at callware.com> writes:

> Hi All--
> 
> Eugene Leitl wrote:
> > 
> > Michael Hudson writes:
> > 
> >  > Last I looked Netscape mailbox format *was* unix mbox format (on unix
> >  > anyway; can't speak for Netscape on windows) - have you tried pointing
> >  > pine at them?
> > 
> > Yes. Both pine and VM. Both see a giant 5 MByte message.
> > 
> >  > Otherwise you could try the mailbox standard library module...
> 
> I remember MHonArc doing OK on Netscape mailboxes, but there were a lot
> of missteps.  You might look into their code to see how they parse the
> mbox into individual messages. ...
> 
> Take care,
> Ivan
> ----------------------------------------------
> Ivan Van Laningham
> Callware Technologies, Inc.
> ivanlan at callware.com
> ivanlan at home.com
> http://www.pauahtun.org
> See also: 
> http://www.foretec.com/python/workshops/1998-11/proceedings.html
> Army Signal Corps:  Cu Chi, Class of '70
> ----------------------------------------------

I once "converted" Netscape mailbox files to VM files. The problem that
I had is different from the problem that is described here, so I think
something else is going on here. Anyway, my problem was that some of the
messages stored in Netscape didn't have an empty line, which VM needed
as a separator between messages, before they began. The effect was that
if I just used Netscape mailbox files in VM, VM would show fewer number
of messages because a few of the adjacent messages were recognized as a
single message. I fixed this by writing a simple shell script, which I
attach below. I suppose it could be useful to somebody else migrating
from Netscape to VM. The only thing that will be necessary to run the
script is to replace `name-temp-file` to some temporary file name. In my
case, name-temp-file is another shell script to generate a new temporary
file name.

--- Cut here for script file 'correct-netscape-mail-file' ---
#!/bin/tcsh -f

onintr cleanup

if ( $# != 1 ) then
    echo usage: $0:t \<mail-file-name\>
    exit 1
endif

set inputfile = "$1"
set tempfile = `name-temp-file`
awk 'BEGIN {prev_empty = 1; bad_lines = 0;} {if ($0 ~ /^From[^:]/) {if (! prev_empty) {bad_lines++; print "";}} print $0; if ($0 == "") prev_empty = 1; else prev_empty = 0;} END {if (! prev_empty) {bad_lines++; print "";} exit (bad_lines);}' "$inputfile" > $tempfile
set nerror = $status
if ( $nerror == 0 ) then
    echo Nothing to correct for mail file \`$inputfile\'.
else
    set backupfile = "$inputfile".backup
    set i = 1
    while ( -e "$backupfile" ) 
	set backupfile = "$backupfile".$i
	@ i = $i + 1
    end
    \mv "$inputfile" "$backupfile"
    echo Backup of the original: \`$backupfile\'
    \mv "$tempfile" "$inputfile"
    echo Number of corrections for \`$inputfile\': $nerror.
endif

cleanup:
    if ( $?tempfile ) then
	if ( -e $tempfile ) \rm $tempfile
    endif
--- Cut here for script file 'correct-netscape-mail-file' ---

-- 
Gey-Hong Gweon
The University of Michigan, Physics Department, Ann Arbor, MI 48109
(734) 647-9434 (Office), 763-3417 (Lab), 763-9694 (Fax) 



From kelvin at scientist.com  Tue Aug 31 22:42:13 1999
From: kelvin at scientist.com (Mr Kelvin Lee)
Date: Wed, 01 Sep 1999 02:42:13 GMT
Subject: language interpreters/ interpreted languages weaknesses?
Message-ID: <7qi3q3$n1d$1@nnrp1.deja.com>

Hi,

The main weakness I hear about language interpreters/ interpreted
languages (btw.. is that the same thing?) is that they are slow(?).. is
this because of the way they handle (data) variables?

Are there any other major weaknesses of interpreted languages over
conventional languages such as C, C++, Ada etc.?

What are the most popular interpreted languages around?

Thanks,

--
Regards,

KL.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From robin at jessikat.demon.co.uk  Mon Aug 16 17:06:59 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Mon, 16 Aug 1999 21:06:59 GMT
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
In-Reply-To: <19990816205639474.AAA315.413@max41127.izone.net.au>
References: <2GJsiHANGbt3Ew6i@jessikat.demon.co.uk>
 <19990816205639474.AAA315.413@max41127.izone.net.au>
Message-ID: <lDvYnDAz1Hu3EwB6@jessikat.demon.co.uk>

In message <19990816205639474.AAA315.413 at max41127.izone.net.au>, John
Machin <sjmachin at lexicon.net> writes
>On 14 Aug 99, at 19:12, Robin Becker wrote:
>
>> I once described perl as being like awk on
>> steroids and lsd
>
>Actually perl is more like SNOBOL4 on sedatives.
>
I still have my snobol4 manual. All sorts of memories with that.
>There is nothing new under the sun. SNOBOL4, which was up to version 3 
>by 1969, had goodies like arrays (like Python lists, but with multiple 
>dimensions that were fixed when the array was created), tables (like 
>Python dictionaries), user-defined functions, user-defined data-types 
...
-- 
Robin Becker



From robin at jessikat.demon.co.uk  Thu Aug 19 14:55:36 1999
From: robin at jessikat.demon.co.uk (Robin Becker)
Date: Thu, 19 Aug 1999 19:55:36 +0100
Subject: Python to Perl Conversions
References: <37bc1303@cs.colorado.edu>
 <slrn7ro9ns.h6v.jblaine@shell2.shore.net> <37bc2ee5@cs.colorado.edu>
 <nJUYxAAwWEv3EwY0@jessikat.demon.co.uk> <37bc505b@cs.colorado.edu>
Message-ID: <9IyORBAoMFv3EwKF@jessikat.demon.co.uk>

In article <37bc505b at cs.colorado.edu>, Tom Christiansen
<tchrist at mox.perl.com> writes
>
...
>
>I never killfiled you.  Why are you trying to start a flame war?  And why
>did you respond with a stupid Jeopardy style Answer before the Question,
>without even trimming?  Is your newsreader broken, or haven't you ever
>bothered to read news.announce.newusers to learn how to use Usenet?
>
>Followups to /dev/null.
>
>--tom
too easy
-- 
Robin Becker



From ScherBi at BAM.com  Thu Aug  5 11:17:07 1999
From: ScherBi at BAM.com (ScherBi at BAM.com)
Date: Thu, 5 Aug 1999 11:17:07 -0400 
Subject: Python on a Novell Netware, Netscape web server
Message-ID: <3F8BA119B48AD11197470008C7A493CB98B4DE@HQOBGEX1>

>>I am working on a port of Python to NetWare as an NLM.
Super cool.

>>Right now this is a NW5 solution
I suppose that a version for NW3.12 is out of the question then, right?

Bill Scherer



From jstevens at bamboo.verinet.com  Sat Aug 14 17:45:04 1999
From: jstevens at bamboo.verinet.com (John Stevens)
Date: Sat, 14 Aug 1999 21:45:04 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu> <37b4d0e0@cs.colorado.edu> <37b6500c.2143627@read.news.global.net.uk>
Message-ID: <slrn7rbor0.6b2.jstevens@bamboo.verinet.com>

On Sat, 14 Aug 1999 11:39:14 GMT, Dave <dave at see.sig.for.address> wrote:
>John, please refrain from cross-posting all your messages to
>comp.lang.python.

I didn't cross post.  I responded to somebody who cross posted.

I never add or remove news groups from the list.

>The 'my language is better than yours' debates are always pointless
>and counterproductive,

Not always.  Such a thread lead me to Python in the first place.

You may be cursing your unlucky stars that that occurred, but
such is life! :-)

>Incidentally you are wrong about perl arrays and python lists being
>different data structures - internally they are both implemented as C
>arrays, only the names are different. 

Which is equivalent to saying that all languages are identical,
because they all end up being machine code, in the end.

Sorry, but that is an over simplification.

>All great ideas start as heresy and end as dogma.

Yes.  The thought that it is actually possible to make objective
measurements and from those measurements determine that one programming
language really *IS* better than another language for a given set
of tasks or environments is, currently (Obviously! ;-> )

HERESY.

;_>

John S



From cbbrowne at news.hex.net  Thu Aug 12 23:25:39 1999
From: cbbrowne at news.hex.net (Christopher Browne)
Date: Fri, 13 Aug 1999 03:25:39 GMT
Subject: Python GUI
References: <7ni69d$nc7$1@nnrp1.deja.com> <37B0E67D.27F8D2ED@prescod.net> <934408001snz@vision25.demon.co.uk> <934485432snz@vision25.demon.co.uk>
Message-ID: <TeMs3.6618$gO1.175350@news2.giganews.com>

On Thu, 12 Aug 99 19:17:12 GMT, Phil Hunt <philh at vision25.demon.co.uk>
wrote: 
>Having looked at Glade, it is now an aim of the Parrot project to
>support its file format.

Cool.  I did a bit of playing with Glade last week, and was similarly
Very Impressed.

Even if it were not useful for building code, it still seems to be a
very nice model of a UI prototyping tool.

"I'd like the screen to look like <this>..."

-- 
Rules of the Evil Overlord #33. "I will never build only one of
anything important.  For the same reason I will always carry at least
two fully loaded weapons at all times." 
cbbrowne at hex.net- <http://www.ntlug.org/~cbbrowne/lsf.html>



From guy_oliver at yahoo.com  Sun Aug 29 11:43:29 1999
From: guy_oliver at yahoo.com (guy_oliver at yahoo.com)
Date: Sun, 29 Aug 1999 15:43:29 GMT
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com> <Pine.LNX.4.04.9908290357540.16114-100000@gnr.u2me3.com> <kdtYkCAgVQy3EwQj@jessikat.demon.co.uk>
Message-ID: <37c94f3a.2285568@news.netset.com>

On Sun, 29 Aug 1999 11:03:12 +0100, Robin Becker
<robin at jessikat.demon.co.uk> wrote:

>>What the hell are you writing that makes the language 'real' to you? Drivers??

>don't be too amazed; the thing that usually makes these languages
>useless is speed. I couldn't write any large scale linear algebra in
>python, perl or java for that matter without using some pretty hefty
>extensions. These are usually carefully crafted in C or Fortran.
>
>The SQL applications are also pretty powerfully assisted. ZOPE has lots
>of pure C.
>
>It's not that it can't be done; it's not fast enough. 

But even these applicaions can benefit from using python as a glue
language.  I cant tell you the number of times I have been asked to
write some program to translate the output of some fortran program
into something that can be feed into the input of a different fortran
program.  If these were wrapped in to modules in Python with a decent
interface, they could be integrated with out even thinking about it.
They certainly wouldnt have to pay me a days worth of programming to
get the two things to work together.  and you'd still get the
performance of using the native C/Fortran program.

Of course Python is too slow for some things, but even in those cases,
it has something to offer.  GUI's, interfaces, integration, and
control programs can be crafted in python to wrap the underlying
C/C++/Fortran programs to make working with these much easier.

I recently was asked to control a data acquisition board from a
program on a pc.  I wouldnt have even thought to try to write the
control software in python, but after I got the control software
written in C, I wrapped it into a python module so I could work with
the thing from in python, where the high level abstractions it offers
are very valuable.  I was able to use the interactive interpreter to
control the system until I had a good understanding of what I could do
with it and then was able to prototype the application in a matter of
hours.  this allowed me to put the app in front of the people that had
requested it much sooner than they had expected, and allowed them to
realize that some of the ideas they had about how it should work were
bogus.  I was able to fix these things just as quickly as I built the
initial app and have a program that is just as fast as the C version,
was done much earlier than expected, and is much closer to what the
users wanted than they would have gotten if I had written it some
other more time consuming way.

I guess what prompted me to start this thread is that I am getting
frustrated by peoples implicit assumption that python is a scripting
language, that scripting languages are for simple programs, and hence,
I'm nuts for suggesting that we use it in some capacity on a given
project that is not simple.  Given enough time, I can convince them
its a good idea, but I have to go through the same 'sales pitch' each
time, bringing out the samples, prototypes, etc that I have to show
what I can do with it, what others have done with it, and educating
them on its abilities.  

The real problem isnt that they think python is a scripting language,
but that they assume that all scripting languages are toys.  Many of
these people are used to thinking of scripting languages as /bin/sh or
batch files on DOS.  And with this mindset, I dont blame them for
questioning my wanting to use a 'scripting' language on a project.
Education is the only solution that there is, but its a slow process.
So as I mentioned in an earlier post, either by separating python from
its scripting heritage, or improving the image of scripting to a point
where its not a bad word among some seems the only quick solution.  Of
course, quick is probably not the right term here.  Both of these are
probably just as time consuming as education.

Thanks to everyone that responded so far!

Guy







From mlindner at mail.com  Wed Aug 25 03:49:24 1999
From: mlindner at mail.com (Michael P Lindner)
Date: Wed, 25 Aug 1999 02:49:24 -0500
Subject: Recommand a good book for learning Python
References: <7pt2oo$rts@netnews.hinet.net>
Message-ID: <7q078v$6nh$1@unlnews.unl.edu>

Alex <do826 at micsv.cwb.gov.tw> wrote in message
news:7pt2oo$rts at netnews.hinet.net...
> Hi all:
>
> I want to learn Python Programming language, but I don't know which book
is
> better for a newbie. So I need some suggestions. Thanks a lot.
>
> P.S: Whould you like to reply one to my E-Mail?
>
> Regards
>
> --Alex

I am using Learning Python by Lutz and Ascher from O'Reilly. So far (through
ch 7) I really like it and have found it easy to understand. I also suggest
buying the Python Pocket Reference from O'Reilly; it's quite handy IMHO.
Mark Lutz also wrote a book called Programming Python, I have heard good
things about it, but don't have it myself as it's possibly dated (1996).

btw: I managed to rewrite a file formatting program I was having trouble
with in C++ the day I bought the book. gl :)

-mike





From ovazquez at sprint.ca  Thu Aug 12 18:42:03 1999
From: ovazquez at sprint.ca (Orlando Vazquez)
Date: Thu, 12 Aug 1999 15:42:03 -0700
Subject: Anyone else find Tk akward?
Message-ID: <37B34DBB.E5A79760@sprint.ca>

Is it just me, or is the way Tk organizes things on the window someone
weird? I'm just starting to play around with it, and it has me puzzled.
Wouldnt it be a lot easier to just use a coordinte system?
For example put a button at (0,0) with length 10 and width 20 etc..
Are there any reasons why it _has_ to be this way?

--
Orlando Vazquez



From bennettt at am.appstate.edu  Tue Aug 31 10:04:36 1999
From: bennettt at am.appstate.edu (TMGB)
Date: Tue, 31 Aug 1999 10:04:36 -0400
Subject: [HOWTO] Launching python from Windows Explorer
References: <s7ca67f3.082@mail.conservation.state.mo.us>
Message-ID: <37CBE0F4.338894BD@am.appstate.edu>

The post below probably could be replicated in Windows 95 and Windows
98.  As an alternative, you can put an executable or shortcut in the
"\windows\sendto" folder.  When you right click on any file you can move
your mouse down to the Send To option and a menu will pop out to the
side listing available applications to "Send to" (ie. all the files in
the sendto folder).  In this way you can creat a shortcut to python.exe
and edit the commandline under properties like the ones below.  I
haven't tested this particularly with python but I have customized the
sendto directory with all executables that I want available from a right
click.  This would even allow you to right click on a python file on the
desk top and run it in case it is something you run often and want
readily available.

Thomas

Jeffrey Kunce wrote:
> 
> I've had several requests to elaborate on an earlier post.
> Perhaps someone could check this, and add to an appropriate FAQ.
> These procedures are for NT, there may be minor differences for Win9x.
> 
> To easily run/open python scripts from the windows explorer,
> I define several "actions" for .py files. Once these actions
> are defined on a machine, you can right-click on a python script,
> and choose the action you want to happen.
> 
> Action Name: test
> Application: "C:\Program Files\Python\python.exe" -i "%1"
> Description: Runs the python script in an MSDOS window. Leaves
> the window open, and the interpreter running, when the script
> finishes. Good for debugging, esp when using the pdb module.
> 
> Action Name: wordpad
> Application: "C:\Program Files\Windows NT\Accessories\wordpad.exe" "%1"
> Description: Opens the python script for editing in wordpad.
> 
> Action Name: idle
> Application: "C:\Program Files\Python\pythonw.exe" "C:\Program Files\Python\Tools\idle\idle.pyw" "%1"
> Description: Opens the python script for editing/running in Idle. Each time
> this action is used, it will open a new instance of Idle. To open additional scripts,
> use Idle's File/Open command.
> 
> To define an action for a ".py" files:
>   -open Windows [NT] Explorer
>   -from the "View" menu, select "Options"
>   -select the "File Types" tab
>   -from "Registered File Types", select "Python File"
>   -click the "Edit" button
>   -click the "New" button
>   -for "Action", enter the action name to be
> displayed in the popup menu
>   -for "Application used", enter the command
> line (as in the examples above)
>   -click "OK", etc...
> 
> Variations of this process will edit existing
> actions, or define actions for new file types.
> 
>   --Jeff

-- 
----------------------------------------------------------------------
Thomas McMillan Grant Bennett           Appalachian State University
Computer Consultant II                  University Library
bennettt at am.appstate.edu               
http://www.library.appstate.edu/admin/
Voice:  828 262 6587			FAX:    828 262 3001

Windows 95 is a 32-bit extension to a 16-bit patch for an 8-bit
operating system that was originally coded for a 4-bit microprocessor.
 - Chris Dunphy     Boot Magazine



From graham at sloth.math.uga.edu  Mon Aug 23 10:28:12 1999
From: graham at sloth.math.uga.edu (Graham Matthews)
Date: 23 Aug 1999 14:28:12 GMT
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <kFHs3.1651$EG4.312151@news.shore.net> <37b34edc@cs.colorado.edu> <37B3DC02.C9DDB210@NOSPAM.strs.co.uk> <37b414d7@cs.colorado.edu> <37c126a6$0$813@newsreader.alink.net>
Message-ID: <7prlps$c05$1@cronkite.cc.uga.edu>

xah (xah at weborder.com) wrote:
: On a personal note...  Do you know what a computer science degree is these
: days? They require you understand shits like lambda calculus, logic, graph
: theory, combinatorics, or even algorithms?? All for what? All the Perl
: programers I know doesn't understand any of these, and they all are making
: big bucks in fortune 500.

And here we have the true meaning of life revealed. The only point in learning
anything is to make big bucks in fortune 500. What a pearl of wisdom that is!

graham
-- 
             So we starve all the teachers
                And recruit more marines
      How come we don't even know what that means
                     Its obvious



From training at eiffel.com  Mon Aug 30 13:33:41 1999
From: training at eiffel.com (Interactive Software Engineering, Inc.)
Date: Mon, 30 Aug 1999 10:33:41 -0700
Subject: Design by Contract Seminar Series
Message-ID: <7qef80$3u7$1@news.rain.org>

ISE'S EXCLUSIVE SEMINARS ON DESIGN BY CONTRACT AND THE COMPONENT REVOLUTION.
A ONE-DAY COURSE BY BERTRAND MEYER.

If you are involved in software development, either as a manager or as a
developer, you know how crucial it is to improve software quality and
productivity. In spite of the major advances of recent years, especially
Object Technology, it is still much too hard to produce correct, reliable
systems at a reasonable cost.

This state of the art seminar is a unique opportunity to hear about the
approach that is revolutionizing the software industry - not from an outside
observer, but from one of the key originators of the discipline, Dr.
Bertrand Meyer, one of a handful of object- oriented movers and shakers
whose work made modern object-oriented technology possible.

"Design by Contract" is the major technological breaktrough in software
engineering in the past 15 years. This course introduces Design by Contract
and shows why it can, like no other technique since the invention of classes
and objects, dramatically affect the way we develop software, and why it is
a required condition for achieving the promise of the Component Revolution.

For more information about the seminar, as well as locations, please visit
our Web site at http://www.designbycontract.com. You can also contact us by
email: training at eiffel.com.

Interactive Software Engineering, Inc
Goleta, California













From donn at u.washington.edu  Tue Aug 10 16:04:39 1999
From: donn at u.washington.edu (Donn Cave)
Date: 10 Aug 1999 20:04:39 GMT
Subject: What's a good way to get my IP from Python?
References: <37B04D5F.2F00EE6C@cisco.com> <2EZr3.1342$EG4.232940@news.shore.net>
Message-ID: <7oq0kn$eii$1@nntp6.u.washington.edu>

"Michael P. Reilly" <arcege at shore.net> writes:
| Aaron Rhodes <aarhodes at cisco.com> wrote:
| : Does anyone know of a really good way to determine the
| : IP address of the ethernet cards on a Linux system
| : using Python only (i.e. no C code/parsing external binaries, etc.) ?
|
| Hopefully this is portable, but my work PC is down and don't have
| access to a Mac; but theoretically it should be.
|
|   import socket
|   def my_ipaddr(interface_hostname=None):
|     # give the hostname of the interface you want the ipaddr of
|     hostname = interface_hostname or socket.gethostname()
|     s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
|     s.bind(hostname, 0)
|     ipaddr, port = s.getsockname()
|     s.close()
|     return ipaddr  # returns 'nnn.nnn.nnn.nnn' (StringType)
|
| You aren't connecting to anything so this _should_ work even if you are
| not on a network at the time (PPP), the interface just needs to be
| plumbed and initialized.  The begged question would be: how do you get
| the hostnames of multiple interfaces from within Python.  Anyone have a
| good answer for that?

I ran into this problem in a PPP situation, and I had to crank harder
than that.  Of course multiple networks is part of the problem (I had
separate local and dial-up networks, both via PPP.)  But for me, even
with only one network, in a dynamic situation like most PPP services,
I got a valid IP only when I attempt to connect to something via that
network.  I more or less arbitrarily chose to connect to my primary
DNS service.

The recent thread on ifconfig interface is germane here.  If the ifconfig
ioctls work on your platform (i.e., UNIX), you can retrieve all kinds
of information about network interfaces, but I don't know if it has any
answer for valid PPP latency worries as discussed above, and at present
it would require some no-portability-guarantees struct packing (see thread
for more.)

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From jack at oratrix.nl  Tue Aug 31 11:32:11 1999
From: jack at oratrix.nl (Jack Jansen)
Date: Tue, 31 Aug 1999 17:32:11 +0200
Subject: GRiNS 1.0 (Multimedia authoring software in Python) available
Message-ID: <19990831153211.C03C035BB1F@snelboot.oratrix.nl>

[Warning: This message could be seen as being commercial]

Folks,
we proudly announce the first commercial release of GRiNS, our multimedia 
authoring and playback system for the internet. It allows you to create SMIL 
presentations (SMIL is the W3C standard for multimedia over the net, and used 
by applications such as RealNetworks G2 player), and completely written in 
Python. Well, completely: there's a few extension modules involved to make 
various media rendering toolkits available, but the whole application 
framework is in Python (And most of the extension modules have found their way 
back into the Python community, or will do so shortly, of course).

Thanks to Python we're able to release GRiNS for all three major platforms at 
the same time: Macintosh, Windows and Unix (SGI and Sun; sorry, no Linux yet, 
due to the fact that we don't know of any decent media libraries (esp. movie 
libraries), so please enlighten us if you know of something).

If you want to have a look at what a Python multiplatform application can look 
like check it out at http://www.oratrix.com/GRiNS . And feel free to ask us, 
Sjoerd Mullender (sjoerd at oratrix.com) or me, if you want to know anything 
about technical details (or commercial details, which we will duly forward:-).

Oh yes: if anyone has good suggestions for places they always use for being 
kept informed about new software (i.e. who we should inform about this 
release) we'd be happy if you could share them with us.

Here's the press release, for completeness sake:
---------------------------------------------------------
Oratrix Development BV

http://www.oratrix.com/

GRiNS for SMIL WILL TRANSFORM MULTIMEDIA WEB AUTHORING

Amsterdam, The Netherlands (August 30, 1999) - Oratrix Development BV today 
announced the release of GRiNS for SMIL V1.0, a unique authoring and 
presentation
system for local and streaming media on the Web.

GRiNS/SMIL allows content providers to create adaptive multimedia productions 
that
tailor content to the needs and resources of the user community. Working 
together with
RealNetworks?? RealSystem? G2, GRiNS provides access to the full capabilities 
of the SMIL
language on Windows, Macintosh and UNIX systems. Users on all these platforms 
can
finally design multimedia presentations with adaptive content.

According to Dick Bulterman, founder and managing director of Oratrix 
Development,
"GRiNS/SMIL will transform Web multimedia from a one-size-fits-all paradigm to 
one in
which the needs, resources and capabilities of individual users can determine 
the quality
and content provided inside a streaming media presentation."

Presentations can be made that not only adjust for bandwidth constraints, but 
also for
language (for example, English and Spanish), or media type (text, image or 
video).

"GRiNS/SMIL fills a critical gap in the authoring market," adds Bulterman. "We 
greatly
simplify the process of creating SMIL presentations for the RealNetworks 
RealPlayer? G2 at
an affordable price." Bulterman, described by UK's Internet Magazine as one of 
the Web's
10 behind-the-sceens heroes, says that the system's ability to create adaptive 
presentations will open up new value-added markets by allowing authors to 
tailor their
content to the needs and resources of their clients.

Commenting on the release, Len Jordan, senior vice-president, Media Systems, 
RealNetworks, Inc. said: "GRiNS provides an intuitive and easy complement to 
Realslideshow(TM) for users who want to exploit multiple media types and 
advanced
capabilities of both SMIL and RealSystem G2." Geoff Freed of CPB/WGBH National 
Center
for Accessible Media added: "The ability of GRiNS/SMIL to provide control of 
adaptive
content for both streaming media and the local GRiNS player will significantly 
lower the
barrier for creating presentations that are accessible to the widest possible 
range of Web
users."  Ivar Brinkman of Elsevier Science Publishers noted: "The fact that 
GRiNS is available
for Macintosh and UNIX as well as Windows-98/NT allows Elsevier to work with 
all of our
authors in evaluating new approaches to showcase and enhance our printed 
publications."

GRiNS/SMIL was developed by Oratrix Development BV, an Amsterdam-based 
spin-off of
CWI, the Dutch national center for mathematics and computer science. Oratrix 
Development is winner of the McKinsey & Co. "New Venture" award and has been 
selected by the Twinning Venture Capital Network in the Netherlands. 
GRiNS/SMIL has
been tested by over two thousand users world-wide in beta form.

GRiNS/SMIL is available at http://www.oratrix.com/GRiNS/.

For more information contact Manja Strick van Linschoten at Oratrix 
Development:
Email: 	Manja.Strickvl at oratrix.com 
TEL: 	+31 20 679 5306
FAX: 	+31 20 679 5309

--
Jack Jansen             | ++++ stop the execution of Mumia Abu-Jamal ++++
Jack.Jansen at oratrix.com | ++++ if you agree copy these lines to your sig ++++
www.oratrix.nl/~jack    | see http://www.xs4all.nl/~tank/spg-l/sigaction.htm 





From dubois1 at llnl.gov  Mon Aug  9 18:05:55 1999
From: dubois1 at llnl.gov (Paul F. Dubois)
Date: Mon, 09 Aug 99 22:05:55 GMT
Subject: Pyfort - Python/Fortran connection tool
Message-ID: <mt2.0-7542-934236355@news.informatik.uni-bonn.de>

Pyfort is a tool for connecting Python to Fortran. Using a syntax similar to
Fortran 95's interface blocks, Pyfort produces a Python module that can be
linked together with the Fortran routines to enable them to be called from
Fortran.

The document is http://xfiles.llnl.gov/pyfort.pdf; it is in Acrobat format.

Pyfort can be downloaded at ftp://ftp-icf.llnl.gov/pub/python/pyfort.tgz. It
is 100% Pure Python; no compilation is required; however, you will need both
C and Fortran compilers for your application.

Pyfort has been released for unlimited redistribution; please see the file
Legal.htm in the distribution.

Paul F. Dubois
Lawrence Livermore National Laboratory
dubois1 at llnl.gov

Questions to support at pcmdi.llnl.gov.


<P><A HREF="http://xfiles.llnl.gov/python.htm">Pyfort 1.0</A> - a tool for
connecting Python to Fortran.  (09-Aug-99)

-- 
----------- comp.lang.python.announce (moderated) ----------
Article Submission Address:  python-announce at python.org
Python Language Home Page:   http://www.python.org/
Python Quick Help Index:     http://www.python.org/Help.html
------------------------------------------------------------



From abigail at delanet.com  Thu Aug 26 11:33:03 1999
From: abigail at delanet.com (Abigail)
Date: 26 Aug 1999 10:33:03 -0500
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <FAA0A36A71C86317.F8AACF1ACF4646E5.12D37B0704A76E00@lp.airnews.net> <37C552DF.9F90BA36@megsinet.net>
Message-ID: <slrn7sanj3.n0r.abigail@alexandra.delanet.com>

David Oppenheimer (davidopp at megsinet.net) wrote on MMCLXXXVI September
MCMXCIII in <URL:news:37C552DF.9F90BA36 at megsinet.net>:
$$ 
$$ That being said, are there ways to integrate Python and Perl to do things
$$ that neither one can do alone?!


Since Python and Perl are Turing equivalent, what things are you refering to?



Abigail
-- 
perl -wle 'print "Prime" if (0 x shift) !~ m 0^\0?$|^(\0\0+?)\1+$0'


  -----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
   http://www.newsfeeds.com       The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including  Dedicated  Binaries Servers ==-----



From roy at popmail.med.nyu.edu  Wed Aug 18 13:06:30 1999
From: roy at popmail.med.nyu.edu (Roy Smith)
Date: Wed, 18 Aug 1999 13:06:30 -0400
Subject: Wacky Programming Tales
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B54DA3.4210DC10@nospam.com> <slrn7rc8co.rsf.sholden@pgrad.cs.usyd.edu.au> <wjy7lmxbwh7.fsf_-_@bradh.ne.mediaone.net> <37b62939@cs.colorado.edu> <slrn7rccem.ne.bmcd@ripley.sulaco.net> <37BAE44B.E66250BD@geo.vu.nl>
Message-ID: <roy-1808991306300001@qwerky.med.nyu.edu>

Patrick Bogaart <bogw at geo.vu.nl> wrote:
> >         i = 0;
> >         while( i < max_index ) {
> >                 /* some copying or update (not modifying i) goes here */
> >                 i = i + 1;
> >         }

> 
>         i = 0;
>         label_1:;
>         /* some use of i */
>         i = i + 1;
>         if (i==max_index) goto label_2;
>         goto label_1;
>         label_2:;
> 
> <wink>

To get the correct behavior for values of max_index less than 0, you need
to change the test from "i == max_index" to "i >= max_index", and you also
need to move the test to directly after label_1.  Other than that, it
looks like perfectly reasonable ratfor-type output (except that the labels
should start at 23001).

-- 
Roy Smith <roy at popmail.med.nyu.edu>
New York University School of Medicine




From jeremy at cnri.reston.va.us  Fri Aug  6 15:22:48 1999
From: jeremy at cnri.reston.va.us (Jeremy Hylton)
Date: Fri, 6 Aug 1999 15:22:48 -0400 (EDT)
Subject: rexec question
In-Reply-To: <xzg11xwf10.fsf@prakinf.tu-ilmenau.de>
References: <xz4sievhzv.fsf@prakinf.tu-ilmenau.de>
	<14249.56828.918361.13395@bitdiddle.cnri.reston.va.us>
	<xzg11xwf10.fsf@prakinf.tu-ilmenau.de>
Message-ID: <14251.13832.844951.142519@bitdiddle.cnri.reston.va.us>

[Just rememberer that the basic of using rexec are explained in Andrew
Kuchling's howto: http://www.python.org/doc/howto/rexec/rexec.html]

>>>>> "RD" == Ralf Doering <ralf.doering at Prakinf.TU-Ilmenau.DE> writes:

  [near miss in my previous attempt to answer the question]

  RD> Thanks for your response. Do yo see any way to get an *already*
  RD> defined object into the namespace of rexec?  The object we talk
  RD> about will be unpickled from a byte stream and has to be
  RD> executed in the restricted environment.  So, how can I get this
  RD> unpickled object into RExec and run it in this namespace?  Seems
  RD> that I`m a little confused about the namespaces and how to
  RD> define entrys in RExec`s namespace ...

I should have answered this question from the outset.  The restricted
execution environment is definitely more interesting when there are
objects defined in the controlling environment that are all accessible 
in the restricted environment.

There is no direct way to add a name binding to the restricted
execution environment.  You need to modify one of the modules that
exists in rexec, typically its __main__.

Here's a simple approach that would work:

class RExecEx(rexec.RExec):
    """Adds method for setting attribute in rexec environment"""
    def setattr(self, name, value, bastion=1):
        if bastion:
            value = Bastion.Bastion(value)
        setattr(self.modules['__main__'], name, value)

In most cases, you want to pass a Bastion to the untrusted code rather 
than passing the object directly.  The howto and the Bastion module
documentation explain this.

I've included a short script that illustrates the damage the untrusted 
code can do if you don't use a bastion.

Jeremy



import rexec
import Bastion 

class Foo:
    _priveleged = 12
    
    def bar(self):
        return self._priveleged

spam = Foo()

untrusted_code = """print "Initial value:", spam.bar()
print "Attempting to read privleged attribute:",
try:
    x = spam._priveleged
    print "succeeded"
except AttributeError:
    print "failed"
    pass
print "Attempting to modify object:",
spam._priveleged = 15
if spam.bar() == 15:
    print "internally modified",
else:
    print "internally unmodified",
if spam._priveleged == 15:
    print "externally modified"
else:
    print "externally unmodified"
print "Final value:", spam.bar()
"""

renv = rexec.RExec()

print "Running user code with direct access to object"
setattr(renv.modules['__main__'], 'spam', spam)
renv.r_exec(untrusted_code)
print "Value outside rexec:", spam.bar()
print

spam = Foo()
print "Running user code with access through Bastion"
setattr(renv.modules['__main__'], 'spam', Bastion.Bastion(spam))
renv.r_exec(untrusted_code)
print "Value outside rexec:", spam.bar()



From frankn=news at cs.vu.nl  Mon Aug  2 08:52:40 1999
From: frankn=news at cs.vu.nl (Frank Niessink)
Date: Mon, 2 Aug 1999 12:52:40 GMT
Subject: IntVar - to detect a change in it
References: <7o3k4f$74n$1@nnrp1.deja.com>
Message-ID: <7o44ao$200@cs.vu.nl>

Ole-Kristian Villaboe <olevi at my-deja.com> wrote:
> Is there anyway an altering of a variable (Int-, String- or Double-)
> (by the user, via the GUI) could call a function in order to tell the
> environment that it has changed since the initial setting(or a static
> comparison value)

Assuming you're talking about Tkinter here: below's an example of using
trace() to trace a variable:

class VLabel(Label):
        """VLabel class - becomes red when textvariable.valid() returns
	false"""

	def __init__(self, master, textvariable, **cnf):
		apply(Label.__init__, (self, master), cnf)
		self.config(textvariable=textvariable)
		self.textvariable = textvariable
		textvariable.trace('w',self.notify)

	def notify(self, *args):
		if self.textvariable.valid():
			self.config(bg=self.config('bg')[3])
		else:
			self.config(bg='red')

Everytime the textvariable is changed notify() gets called, which checks
whether the textvariable is valid. If not, the background is colored red,
if it is valid, the background is set to its default color.

HTH, Frank
-- 
The last ever dolphin message was misinterpreted as a surprisingly
sophisticated attempt to do a double-backwards-somersault through a hoop 
whilst whistling the "Star Sprangled Banner", but in fact the message was 
this: So long and thanks for all the fish.
	-- Douglas Adams, 'The Hitch Hiker's Guide to the Galaxy'



From tchrist at mox.perl.com  Wed Aug 18 23:17:07 1999
From: tchrist at mox.perl.com (Tom Christiansen)
Date: 18 Aug 1999 20:17:07 -0700
Subject: Compound Assignment Operators ( +=, *=, etc...)
References: <37B47506.14EBD269@msfc.nasa.gov> <c75yafced1p.fsf@pc142.cosc.canterbury.ac.nz>
Message-ID: <37bb6923@cs.colorado.edu>

     [courtesy cc of this posting mailed to cited author]

In comp.lang.python, 
    Timothy R Evans <tre17 at pc142.cosc.canterbury.ac.nz> writes:
:One reason that these constructs could cause problems is due to
:namespaces.  Assuming that x += y is expanded to x = x + y, consider
:the following code while remembering how python treats namespaces and
:global variables.
:
:>>> def foo():
:...    x += 2
:...    return x
:>>> x = 42
:>>> foo()
:44
:>>> x
:42
:
:I don't think this is what you want to happen, but it is what python
:would do.  Inside foo(), the global variable x is read, then a local
:variable x is created (there is no `global x' in the function).  So in 
:this case x += 2 does not add 2 to x, it creates a new local variable
:that is 2 greater than the global variable of the same name.

That's not the way I dimly understand it.  You can't from a local
scope even try to do a 

    x = x + 2

on a global, unless you declare x to be global.  Anything else is illegal,
because you'd end up talking about a global and a local at the same
time.  Consider if your function were written:

    def foo():
	x = x + 2
	return x

That's not legal even now.  You need the global.

:Confused yet, well here's a second reason.
:
:>>> x = [1,2,3,4]
:>>> y = x
:>>> x += [5,6]
:>>> x
:[1,2,3,4,5,6]
:>>> y
:what should this be?
:
:you would expect x += y on lists to actually extend the list, or do
:you want it to be x = x + y, which could be really slow on big lists,
:while hiding the fact that it is copying the list.

I think you've pointed out a real issue here, but it's not the "big lists
are slow one".  That's just something the compiler should just optimize.

The real issue is the aliasing issue.  If you write:

    x = [1,2,3,4]
    y = x

Then you are expecting x and y to refer to the same object.  If you use

    x = x + [5,6]

then you've orphaned y, which remains as it was due to the
semantics of "+" providing a new result list.

Is this bad?  I don't know for sure.  Sometimes it seems to make sense.
Certainly if you write

    x = 4
    y = x
    x = x + 2

you don't expect y to track the changes to x and also be 6.  

I don't really see any reason for 

    var += expr

to be defined in any terms other than

    var = var + expr

However, I do see ample opportunity for optimization, some of which
might be easier given += then it would be if written out long hand.

--tom
-- 
    "The purpose of most computer languages is to lengthen your resume by
    a word and a comma." --Larry Wall



From skip at mojam.com  Tue Aug  3 16:54:36 1999
From: skip at mojam.com (Skip Montanaro)
Date: Tue,  3 Aug 1999 15:54:36 -0500 (CDT)
Subject: Any clues to source of this delay?
In-Reply-To: <al8vhaxtg1y.fsf@sirppi.helsinki.fi>
References: <al8vhaxtg1y.fsf@sirppi.helsinki.fi>
Message-ID: <14247.22099.356207.503879@dolphin.mojam.com>

    Markus> Problem:

    Markus> On Linux, single connections have 'glass ceiling' of 50
    Markus> roundtrips/second; no matter what I kludge, it seems to stay
    Markus> there. Only way to fix this was to send 'empty' packets after
    Markus> processed ones, thus increasing throughput by about 5X.

    Markus> On NT, the glass ceiling is same, just 1/10 of Linux's (5
    Markus> roundtrips/second).

TCP's slow start algorithm, perhaps?  If you can figure a way to reuse the
connection for multiple messages you may find that subsequent messages are
sent more quickly.  A quick check at Google suggests that slow start details 
might be found in RFC's 2001 and 2581.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-475-3758



From Dan at Grassi.com  Sat Aug 14 20:30:01 1999
From: Dan at Grassi.com (Dan)
Date: Sat, 14 Aug 1999 20:30:01 -0400
Subject: Mac ODBC
Message-ID: <199908150028.RAA26649@mb3.mailbank.com>

Hi M.-A. Lemburg,

>Does the driver expose the ODBC API in a way that allows you to
>link against it ? If yes, then you could try to create a new
>subpackage for mxODBC (see my Python Pages for more infos).

Yes it does.  I initially did not think that it did, a co-worker had 
installed the ODBC package and when I asked him he said that no source 
files were included.  Your message got me thinking and checking and yes 
there were files and libraries.

When I get this working I will let you know and make the project source 
available.

Thanks,

Dan



From nascheme at ucalgary.ca  Thu Aug 19 19:14:20 1999
From: nascheme at ucalgary.ca (Neil Schemenauer)
Date: 19 Aug 1999 23:14:20 GMT
Subject: ANN: mxProxy Package - Version 0.2.0
References: <37BC70EB.7E03BBEB@lemburg.com>
Message-ID: <7pi34c$cgg@ds2.acs.ucalgary.ca>

M.-A. Lemburg <mal at lemburg.com> wrote:
>2. using a weak reference which allows you to build circular
>   reference models in your data without producing the other-
>   wise difficult to handle memory leakage induced by this.

This is an interesting implementation of weak references.
Thanks.  Attached is a weak reference implementation I cooked up
yesterday.  It is more of a hack then your approach but has the
advantage that referenced objects die sooner.  Unfortunately, my
proxy object is incomplete.

    Neil


"""
Weak References in Python

This is a hack.  It works okay in the current version of CPython.  Any object
that has settable attributes can have weak references to it.  To get a weak
reference to an object use WeakRef(object).  When the object is destroyed any
attempts to access it though the weak reference will raise an exception.  

This module requires the refhack extension module, available from:

    http://www.ucalgary.ca/~nascheme/python/

Neil Schemenauer <nascheme at ucalgary.ca>
August, 1999
"""

import refhack

# something that cannot conflict with normal Python code
MAGIC_NAME = ' __weakref_cleaner'

WeakRefError = 'WeakRefError'

class WeakRef:
    """Create a weak reference and return a proxy object"""
    def __init__(self, obj):
        self._alive = 1
        self._obj = obj
        if not hasattr(self._obj, MAGIC_NAME):
            cleaner = WeakRefCleaner(self._obj)
        else:
            cleaner = getattr(self._obj, MAGIC_NAME)
        cleaner.add_ref(self)
        refhack.decref(self._obj)

    def _destroy(self):
        if self._alive:
            # _obj is now gone, don't touch it.  The refcount of _obj will go
            # negitive but this is okay in the current Python implementation
            self._alive = 0

    def __getattr__(self, name):
        if not self._alive:
            raise WeakRefError, 'object has been destroyed'
        return getattr(self.__dict__['_obj'], name)

    #def __setattr__(self, name, value):
    #    if not self._alive:
    #        raise WeakRefError, 'object has been destroyed'
    #    self.__dict__['_obj'].setattr(name, value)


class WeakRefCleaner:
    """A little object to die with the weak referenced object.  It
    will inform all weak references of the objects death"""
    def __init__(self, obj):
        self.refs = []
        # the object must have settable attributes
        setattr(obj, MAGIC_NAME, self)

    def add_ref(self, ref):
        self.refs.append(ref)

    def remove_ref(self, ref):
        self.refs.remove(ref)

    def __del__(self):
        for ref in self.refs:
            ref._destroy()


def test():
    class A:
        pass
    a = A()
    a.b = 1
    r = WeakRef(a)
    print r.b
    del a
    print r.b

if __name__ == '__main__':
    test()



From kistler at fnmail.com  Sat Aug 21 05:23:31 1999
From: kistler at fnmail.com (Per Kistler)
Date: Sat, 21 Aug 1999 11:23:31 +0200
Subject: UnboundLocalError (was RE: Compound Assignment Operators ( +=, *=, 
 etc...))
References: <000101beebac$75da5de0$0c2d2399@tim>
Message-ID: <37BE7013.C2DA2200@fnmail.com>

Hi Tim

Could you explain to me the meaning of "<wink>", please?
Is it a form of ":-)"?

-Per.

Tim Peters wrote:

... <wink>.

-- 
Per Kistler kistler at fnmail.com / kistler at gmx.net
------------------------------------------------------------



From digitome at iol.ie  Sat Aug  7 06:20:16 1999
From: digitome at iol.ie (Sean Mc Grath)
Date: Sat, 07 Aug 1999 10:20:16 GMT
Subject: Stupid question about .EXE
References: <7o93uk$sjrn$1@reader2.wxs.nl> <7ob01f$ghn$1@nnrp1.deja.com> <GN8q3.543$L73.32637041@news.randori.com>
Message-ID: <37ac072f.2341316@news.iol.ie>

[Robin Dunn]
>Also take a look at Grodon McMillan's Installer utility found at
>http://starship.python.net/crew/gmcm/install.html.  This one doesn't require
>a C compiler but is Win32 only (and works quite good in my opinion.)
>

I second that! We have just finished a freeware utility to convert
Eudora mailboxes to XML and from there to the Folio Views
IR engine. The thing is written in Python using wxPython as
the front end and shipped as an installshield setup. Gordon's
installer did the magic of accumulating all the dependent
bits into one directory and creating the win32 executable.

Nobody who has used it so far has been able to guess that it
was written in Python. For fun, we are winding up a few
people, telling them that we took the Microsoft C++ source
code for the Scribble application as a base...





From catlee at my-deja.com  Mon Aug  9 17:55:46 1999
From: catlee at my-deja.com (catlee at my-deja.com)
Date: Mon, 09 Aug 1999 21:55:46 GMT
Subject: gzip bug?
Message-ID: <7onios$4dt$1@nnrp1.deja.com>

gzip file objects don't seem to close themselves automatically.  When I
do this:
cPickle.dump(object,gzip.open('file','w'))

Only a couple hundred bytes are written.  However everything works find
when I do this:
f=gzip.open('file','w')
cPickle.dump(object,f)
f.close()

Cheers,
Chris


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From philh at vision25.demon.co.uk  Tue Aug 24 15:28:10 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 24 Aug 99 19:28:10 GMT
Subject: *** Parrot-0.0.1 released ***
References: <935279776snz@vision25.demon.co.uk> <935454399snz@vision25.demon.co.uk> <37C2B9CE.95AA5FEC@lri.fr>
Message-ID: <935522890snz@vision25.demon.co.uk>

In article <37C2B9CE.95AA5FEC at lri.fr> conversy at lri.fr "Stephane Conversy" writes:
> Could you give an example of how to plug code with a generated python
> file ?

Not in detail, because I haven't written a Python backend yet.
 
> for example, how to attach a command when the "New" menu is triggerred ?
> 
> window @MyWindow "My First Window" {
>      menuBar {
>         menu "File" {
>            menuItem @New "New"
>         }
> }

The way I envisage this working is that it would create a class,
MyWindow, for the window. This would have an empty method called
New() or New_pressed() or somesuch. To write code for this method, you
would subclass the MyWindow class with a class that implemented
the method.
 
(The details of this are no doubt wrong, but that is the general scheme
I have in mind).

-- 
Phil Hunt....philh at vision25.demon.co.uk




From chad at vision.arc.nasa.gov  Fri Aug 20 17:07:29 1999
From: chad at vision.arc.nasa.gov (Chad Netzer)
Date: Fri, 20 Aug 1999 21:07:29 +0000
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
References: <Pine.LNX.4.10.9908200706500.15847-100000@starship.python.net>
Message-ID: <37BDC391.83248D8B@vision.arc.nasa.gov>

Moshe Zadka wrote:

> Make that:
> Python: Programming the way g--d- indented.

Ha ha!  Now that's clever!

Chad





From skip at mojam.com  Wed Aug 18 14:49:48 1999
From: skip at mojam.com (Skip Montanaro)
Date: Wed, 18 Aug 1999 13:49:48 -0500
Subject: "connection refused" error from UDP server???
Message-ID: <199908181849.NAA16763@dolphin.mojam.com>

I'm running a simple UDP echo server based on the sample in the Python
distribution:

    def server(port, verbose=0):
	signal.signal(signal.SIGHUP, exithandler)
	s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
	s.bind('', port)
	while 1:
	    data, addr = s.recvfrom(BUFSIZE)
	    if verbose: sys.stdout.write(data)
	    s.sendto(data, addr)

When I run it, I get a "connection refused" message on the recvfrom call
when it's pinged:

    >>> import udpecho
    >>> udpecho.server(10417, 1)


    Traceback (innermost last):
      File "<stdin>", line 1, in ?
      File "/usr/local/lib/automatrix/python/udpecho.py", line 53, in server
	data, addr = s.recvfrom(BUFSIZE)
    socket.error: (111, 'Connection refused')

even though on the other end, things seem to complete okay:

    >>> sockmisc.ping("localhost", 10417)
    ?? ('\012', ('127.0.0.1', 10417))
    0.01

(The line beginning "??" echoes the response from the server and its
address.)

What gives?

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098



From greg at cosc.canterbury.ac.nz  Wed Aug 11 21:51:41 1999
From: greg at cosc.canterbury.ac.nz (Greg Ewing)
Date: Thu, 12 Aug 1999 13:51:41 +1200
Subject: Tkinter newbie question
References: <cThs3.135$Vn5.17296@newsfeed.avtel.net>
Message-ID: <37B228AD.A282DE30@cosc.canterbury.ac.nz>

Clarence Gardner wrote:
> 
> So, is there a general way to know what to do after you've found a Tk
> function that you want to call?

Tkinter is a wrapper around the Tcl interface to
Tk, not the C interface. So looking in the C
interface documentation is generally not helpful.

The traditional way is first to work out how you 
would do it from Tcl (using the Tcl-level documentation 
for Tk) then find out how that interface has been 
translated into Python (by reading Tkinter.py).

There is also some alternative Tkinter documentation 
available now, written with less reference to Tcl terms. 
See

http://starship.python.net/crew/fredrik/tkintro/Introduction.htm
http://starship.python.net/crew/fredrik/tkclass/

It's not complete enough to completely replace the
Tcl documentation, however.

To answer your specific question, I think you need
to use the geometry method of the window:

	win.geometry(newGeometry)

which corresponds to the Tcl "wm geometry" command.

-- 
Greg Ewing, Computer Science Dept, | The address below is not spam-
University of Canterbury,	   | protected, so as not to waste
Christchurch, New Zealand	   | the time of Guido van Rossum.
greg at cosc.canterbury.ac.nz



From tim_one at email.msn.com  Sun Aug 29 20:41:07 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sun, 29 Aug 1999 20:41:07 -0400
Subject: fo.flush() & fo.write() on windows
In-Reply-To: <37C9C25F.DE0BEEAB@rl.gov>
Message-ID: <000001bef280$597f4040$a22d153f@tim>

[Roberts, Robert J]
> Am I asking too much?

On Windows, yes.  C's stdio isn't its native file-handling system, and its
implementation of stdio is, umm, "surprising".

> pseudo code:

>     fo = __builtin__.open(foName, 'a')

Why not plain "open"?

>     foSize = os.stat(foName)[6]     # if foSize = 100...

os.path.getsize(foName) is easier on the brain (new in 1.5.2 though).

>     fo.write('Some string.')             # fo.write() returns None

file.write() always returns None, as documented.  Don't know what you expected
it to return, but read the docs.

>     foSize = os.stat(foName)[6]    # ...foSize STILL = 100.

Unsurprising so far.

>     fo.flush()
>     foSize = os.stat(foName)[6]    # ...foSize STILL = 100.

Ah -- *that's* Windows <wink>!

>     fo.close()                     # 'Some string' appears in fo.
>     foSize = os.stat(foName)[6]    # [now the bytes appear]
>
> I am running Python 1.5.2 on Windows 95. fo.write() appears to be
> buffering the write and returning None for bytes written on
> fo.write().

"None" is expected (see above).  Buffering should be expected too since you
opened the file with buffering.  If you don't want buffering, pass 0 as the
third arg:

     fo = open(foName, 'a', 0)

Alas, on Windows that won't make much of a difference in what you see.

> fo.flush() appears to do nothing.

MS's implementation of stdio's fflush (which Python's file.flush() calls
directly) does clear the stdio buffers, but merely flushes them to Windows'
internal buffers.  It does not imply commitment to disk.  The only way to get
that under Windows is to pass the non-standard "c" (for Commit) mode flag to
"open":

    fo = open(foName, 'ac')

Then fo.write(whatever) *still* won't (necessarily) show up on disk right away,
but fo.flush() will work the way you expect.  "c" is unportable, and much as
you may think you want it now <wink>, slows down Windows output enormously.

windows-was-designed-under-the-assumption-that-it-would-
    never-crash<snort>-ly y'rs  - tim





From ovazquez at sprint.ca  Thu Aug 12 01:34:11 1999
From: ovazquez at sprint.ca (Orlando Vazquez)
Date: Wed, 11 Aug 1999 22:34:11 -0700
Subject: configure is screwing up threadmodule.c
Message-ID: <37B25CD3.E25A4CC5@sprint.ca>

hi, i'm having problems getting threads to work in python. my problem is
at compile time. i have pthreads 1.0 installed on my linux computer
(slackware 3.4/kernel 2.0.30/gcc v2.7.2.3) when i ./configure python i
pass it the --with-threads. maybe something is going wrong i'm not sure
as to what to look out for?

Sorry about the lack of 'proof' so to speak. there is a lot of stuff in
the output of configure, and all the makes.. if you can give me any
advice as to what i should show you please tell me! i found this bit in
the configure script.

Any ideas/comments/suggestions as to what's happening here? I really
needs threads right now.
...
checking for pthread.h... (cached) yes
...
checking for --with-dec-threads... no
checking for --with-threads... yes
checking for --with-thread... yes
checking for mach/cthreads.h... (cached) no
checking for pthread_detach... (cached) no
checking for kernel/OS.h... (cached) no
checking for pthread_create in -lpthreads... (cached) no
checking for pthread_create in -lpthread... (cached) no
checking for pthread_create in -lc_r... (cached) no
checking for __d6_pthread_create in -lthread... (cached) no
checking for pthread_create in -lcma... (cached) no
checking for usconfig in -lmpc... (cached) no
checking for thr_create in -lthread... (cached) no
...



From tim_one at email.msn.com  Sat Aug 14 23:09:48 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Sat, 14 Aug 1999 23:09:48 -0400
Subject: Compound Assignment Operators ( +=, *=, etc...)
In-Reply-To: <37b5a8e0@cs.colorado.edu>
Message-ID: <000501bee6cb$a31710e0$832d2399@tim>

[courtesy cc of this posting mailed to cited author]

[Drew McDowell]
> I looked in the FAQ and couldn't find anything on the Compound
> Assignment Operators.  (+=, *=, -=, etc..)  Why doesn't Python
> support them?

Look in the ToDo list, or search DejaNews very hard <wink>.

> Is there a easy way to add them?

Nope.

[switching to Tom Christiansen]
> Well, to start with, Python does not even *have* an assignment
> operator.  Nope, not one.  Assignment is a statement, not an
> operator.
> [followed by the familiar confusions owing to assignment operators
>  in C]

That's indeed why Python doesn't have assignment operators.  Drew was really
asking why Python doesn't have Compound Assignment Statements (he'll get
back to why it doesn't have assignment operators  ...), so:

> ...
> As for why you can't write the statement (note: not the expression)
>
>     x["every"]["good"]["boy"]["does"]["fine"] =  (
> 	x["every"]["good"]["boy"]["does"]["fine"] + 1 )
>
> as
>
>     x["every"]["good"]["boy"]["does"]["fine"] += 1
>
> The basic reason seems to be that there are some people who believe
> that this kind of thing is cruel and unusual punishment levied upon
> the poor Pascal(ish) programmers purely for sake of brevity.

THere are, but Guido isn't one of them.  He's on record as liking augmented
assignments well enough that he'd put them in someday if more than 6 people
cared <wink>.  Seriously, the problem at the start was simply one of
non-obviousness.  For example, if x and y are lists, does

    z = x
    x += y
    print z

mean

    x = x + y   # x is now bound to an entirely different object
    # and the old x is printed

or

   x.extend(y)   # x is still the same object, but mutated
   # and the new x is printed

?  The resulting *values* compare equal either way, but there are
differences in efficiency and effects on object identity.

There were more important questions to resolve first, and the lack of
augmented assignments hasn't caused enough people enough pain to force a
resolution of these questions.  Nevertheless, in good Pythonic fashion,
Guido's favored resolution is to say it's up to the object:  whatever type x
may happen to have, it's up to x.__addeq__(y) to decide what "+=" (& so on)
means.

> Common arguments in support of such ops include:
>
>     1) This requires care to avoid mis-duplication:
> 	...
>     2) Side effects must be duplicated:
>     ...
>     3) It's too long to type.
>     4) Needless redundancy obscures the real meaning.

5) Efficiency.

> Whereas common rebuttals include:
>
>     1) Toss /bin/ed and get yourself a real editor.
>     2) Don't use side-effects, you wicked creature, you!
>     3) It's more efficient to keep a tmpobj anyway.
>     4) Redundancy can clarify through redundancy.

5) Not having it was good enough for my grandfather, and what makes you
think you're better than him <wink>?

> ...
> But the best reason is probably the one another poster already
> supplied: because Guido didn't like them.

Not so in this case (++x and x++ are a different matter).  Supply & demand,
and need vs desire, get closer.

> An interesting study would be to check the extent to which he
> also forbids such constructs in his own C code. :-)

Python's implementation begins with:

#define colon {
#define dedent }
#define spaces(E) (E)
#define rspace )
#define newline ;

if spaces(sVersion == 0) colon
    printf("internal error") newline
    exit(1) newline
    dedent

if-perl-really-believed-in-freedom-it-would-have-a-
    character-level-preprocessor<wink>-ly y'rs  - tim





From ivanlan at callware.com  Fri Aug 27 08:32:36 1999
From: ivanlan at callware.com (Ivan Van Laningham)
Date: Fri, 27 Aug 1999 12:32:36 GMT
Subject: A Plea For A Return To Sanity!!!
References: <000001bef041$84deffa0$592d153f@tim>
Message-ID: <37C68564.15207535@callware.com>

Hello, All--

Tim Peters wrote:
> 
> [Ivan Van Laningham]
> > 1)  If the Perl yahoos would quit cross-posting flaming bags of shit,
> > it'd be over.
> 
> Actually, I'm afraid my reply here (as well as yours) keeps it going.
> 

I beg to differ.  I didn't cross-post, and your reply wasn't xposted
either.  You and I and all the other Pythonistas are being reasonable,
inoffensive and sane, as always!-)  I suspect that if you could somehow
quantify "flaming bag of shit"-ness, you would find a correlation
between that and cross-posting.

[snip]

> If they're as funny as Mr. X, we can only hope.  You should read Xah's posts
> again!  They're a good parody of the worst of "Usenet Attitude" postings.  We
> may have a potential Python convert there <wink>.
> 

OK, I'll try to read them.  But if you hear a distant explosion, it is
the top of my head blowing off in SLC.

>
> 
> Well, I read all of 'em, and they really were mostly about Python. 

I confess to being unable to read all of them all the way through.  I
read lots of TC's postings, however, and again I must differ with you;
they *pretended* to be mostly about Python.  What they were about,
however, was how miserably Python failed to meet the exquisitely perfect
standards of Perl.  Charitably, this is sandbagging.  Not so charitably,
it makes one wonder if perhaps TC is running for public office. ...

> Besides,
> Tom's very public misunderstandings are a *great* base on which to build a
> "Python for Perlers" FAQ.  Python isn't self-evident at first glance, and it's
> good to relive the initial confusions through his articulate eyes <ahem> (you
> may not have liked how he phrased things, but there was no doubt about what he
> was saying -- that's good writing).
> 

Oh, I agree with this.  Let us cast our Pythons before the perls. 
(Swine are actually highly intelligent, and their disdain for pearls
simply an indication of their pragmatic world-view.)

Quite seriously, I think this is a very good idea.  If I knew more than
nothing about Perl, I would volunteer to put such FAQ together; but not
only do I know zilch, I don't even have time to be writing this
letter;-)  I suggest that if such a FAQ project is picked up by someone,
we leave out the abrasive comments by TC regarding the assholes here
among the Pythonistas.  Why warn potential converts when we can flame
them?

> or-is-this-a-competition-to-see-who-has-the-thinnest-skin?-ly y'rs  - tim

<i-win-i-win!>-ly y'rs,
Ivan
----------------------------------------------
Ivan Van Laningham
Callware Technologies, Inc.
ivanlan at callware.com
ivanlan at home.com
http://www.pauahtun.org
See also: 
http://www.foretec.com/python/workshops/1998-11/proceedings.html
Army Signal Corps:  Cu Chi, Class of '70
----------------------------------------------



From mal at lemburg.com  Sun Aug 29 05:51:13 1999
From: mal at lemburg.com (M.-A. Lemburg)
Date: Sun, 29 Aug 1999 11:51:13 +0200
Subject: Loading shared modules from an imbedded interpreter
References: <37C84727.4E1EB9AF@delanet.com>
Message-ID: <37C90291.BC3E99@lemburg.com>

Dave Wallace wrote:
> 
> Hello,
> 
>     I am attempting to embed Python in a c executable, unfortunately
> when I attempt to import a shared-c module or a Python module that
> imports a shared-c module, I get errors like:
> 
>   Traceback (innermost last):
>   File "<string>", line 1, in ?
> ImportError: /usr/local/lib/python1.5/lib-dynload/selectmodule.so:
> undefined symbol: _Py_NoneStruct
> 
> where the undefined symbol varies from file to file. Apparently, when
> compiling parts of libpython are left out since they are not referenced
> by my code. Is there any way to force them to be included?

You'll have to add an option to the linker flags used to link
Python. These of course vary depending on the linker used. For
GNU ld it is (copied from `man ld`):

       -E

       -export-dynamic
              When  creating  an ELF file, add all symbols to the
              dynamic symbol table.  Normally, the dynamic symbol
              table contains only symbols which are used by a dy?
              namic object.  This option is needed for some  uses
              of dlopen.

If that doesn't work, try to change this part in Python/importdl.c:

#ifdef RTLD_NOW
		/* RTLD_NOW: resolve externals now
		   (i.e. core dump now if some are missing) */
		void *handle = dlopen(pathname, RTLD_NOW);
#else
		void *handle;
		if (Py_VerboseFlag)
			printf("dlopen(\"%s\", %d);\n", pathname,
			       RTLD_LAZY);
		handle = dlopen(pathname, RTLD_LAZY);
#endif /* RTLD_NOW */
		if (handle == NULL) {
			PyErr_SetString(PyExc_ImportError, dlerror());
			return NULL;
		}
		if (fp != NULL && nhandles < 128)
			handles[nhandles++].handle = handle;
		p = (dl_funcptr) dlsym(handle, funcname);

...to the old version (pre 1.5.2):

#ifdef RTLD_NOW
		/* RTLD_NOW: resolve externals now
		   (i.e. core dump now if some are missing) */
		void *handle = dlopen(pathname, RTLD_NOW | RTLD_GLOBAL);
#else
		void *handle;
		if (Py_VerboseFlag)
			printf("dlopen(\"%s\", %d);\n", pathname,
			       RTLD_LAZY | RTLD_GLOBAL);
		handle = dlopen(pathname, RTLD_LAZY | RTLD_GLOBAL);
#endif /* RTLD_NOW */
		if (handle == NULL) {
			PyErr_SetString(PyExc_ImportError, dlerror());
			return NULL;
		}
		if (fp != NULL && nhandles < 128)
			handles[nhandles++].handle = handle;
		p = (dl_funcptr) dlsym(handle, funcname);

Note the use of the dlopen flag RTLD_GLOBAL. This flag was dropped
for 1.5.2 due to various platform dependent problems. Perhaps
we should readd the flag depending on a configure switch...

-- 
Marc-Andre Lemburg
______________________________________________________________________
Y2000:                                                   127 days left
Business:                                      http://www.lemburg.com/
Python Pages:                           http://www.lemburg.com/python/




From guido at cnri.reston.va.us  Fri Aug 20 08:27:16 1999
From: guido at cnri.reston.va.us (Guido van Rossum)
Date: 20 Aug 1999 08:27:16 -0400
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk> <37BC8D05.142E1DB0@aw.sgi.com> <37BC9CE8.67097FC6@vision.arc.nasa.gov> <37BCD6CF.B7A186BC@montana.com>
Message-ID: <5lzozmskm3.fsf@eric.cnri.reston.va.us>

bowman <bowman at montana.com> writes:

> Chad Netzer wrote:
> > 
> > PS.  What is the way to register a callback so that you can intercept the
> > window manager's request to destroy a window?  Then you could just bind
> > the sys.exit call to this on the Tk object...
> 
> def main():
> 	global top
> 	top = Tk()
> 	top.withdraw()
> 	
> 	mb = buildMenubar()
> 	mb.entryconfigure(2, activebackground='red')
> 	myTop = Toplevel(menu=mb)
> 	myTop.bind('<Destroy>', dying)
> 
> 	
> def dying(event) :
> 	print 'Toplevel is getting destroyed'
> 	top.quit()
> 
> 
> The above snippet is a little ugly, but it works. Otherwise, if you kill
> the visible window with the system box, the top process keeps on
> chugging along. I like the Toplevel's menu setup rather than using
> Menubutton's and the deprecated (atleast in the Tcl/Tk world)
> tk_menuBar.

Actually, the accepted idiom (all over the place in code I write) is

        toplevel_window.protocol("WM_DELETE_WINDOW", close_callback)

The close_callback function takes no arguments.  In simple cases, you
can use

	root.protocol("WM_DELETE_WINDOW", root.destroy)

or

	root.protocol("WM_DELETE_WINDOW", root.quit)

--Guido van Rossum (home page: http://www.python.org/~guido/)



From connolly at w3.org  Tue Aug 17 04:34:34 1999
From: connolly at w3.org (Dan Connolly)
Date: Tue, 17 Aug 1999 08:34:34 GMT
Subject: if you want POSIX, import posix, not os
References: <37B8EC03.835F8A51@w3.org> <14265.5096.537800.149227@buffalo.fnal.gov>
Message-ID: <37B91E9A.2451936C@w3.org>

Charles G Waldman wrote:
> 
> Dan Connolly writes:
>  > I have thought this was broken for years, but I just read
>  > it again tonight, and I'm not going to pass over it in silence this
>  > time:
>  >
>  > "Do not import this module directly. Instead, import the module os,
>  > which provides a portable version of this interface."
>  >      -- http://www.python.org/doc/1.5.2p1/lib/module-posix.html
>  >
>  > I think this is terrible advice.
> 
> I think it's good advice.  If you're not using features like
> "getpgrp", but more mundane things like "listdir", then
> 
> import os
> files=os.listdir(os.getcwd())
> 
> is highly portable (across Mac, Win, and Unix) whereas if you spelled
> it
> 
> import posix
> files=posix.listdir(posix.getcwd())
> 
> you'd be introducing a needlessly non-portable construction in your
> code.

But that's not what it says. It says that even if you want
to call stuff that's only available on POSIX platforms,
you should import os rather than posix.

Let's have a module with things like listdir
and getcwd in it, and leave getpwent() completely out of it.

Or at least: document getpwent() under posix, and make a
note by getcwd(), listdir() etc. that says "these are also available
via the os module, which is portable."

> It's up to you to read the documentation to know what are the
> "generic" (always-available) operating system services and which ones
> are platform-specific.  The "getpgrp" example you cite is one of the
> latter...  if you use it, you  won't be able to get your code to run
> on non-Unix platforms.  You have to resort to things like:
> 
> import os
> if os.name = 'posix':
>    pgrp = os.getpgrp()
> else:
>    #something completely different

Why is that better than:

try:
  import posix
except:
  # something completely different

Your code example reminds me of pre-OOP code like:
	if shapeName='circle':
		... draw circle on canvas ...
	elif shapeName='square':
		... draw square
	...


>  > The os module is *not* portable:
>  >
>  > "getpgrp ()
>  >     Return the current process group id. Availability: Unix. "
>  >      -- http://www.python.org/doc/1.5.2p1/lib/os-procinfo.html
> 
> As the "Availability" comment implies, the os module implements a core
> set of functions, on top of which there are platform-specific
> extensions.  If you're not on the right platform, you don't get the
> extensions.

Why glom this into one module? What's the benefit over
having aa well-defined cores set of functions in one module,
and extensions in modules like posix, Win32, etc?

Most of all: why document getpwent() under the os module rather
than under posix?

>  > A portable interface is one where I can expect consistent behaviour
>  > of named objects across platforms, no?
> 
> Modulo the restrictions of "Availability"

Why mess with availability at all? What does it help?


>  > The IEEE POSIX spec is widely implemented and understood.
>  > When a program expects behaviour as specified in the POSIX
>  > specs, it should say
>  >      import posix
> 
> If you make heavy use of the posix features this may make sense.  If
> all you really need is things like listdir, getcwd, then it's
> overkill.
> 
>  > It would have made sense for python to define a really portable os
>  > module, i.e. the least common denominator of the platforms it
>  > runs on. But that's not what we've got.
> 
> No reason you can't create an "generic_os" module something like this:
> 
> ----generic_os.py-----
> from os import listdir,getcwd,path,name,remove,....
> ----------------------
> 
> replacing "...." with the list of all functions which are supported on
> ALL OS's where Python runs (or all OS's you care about).  A python
> program which only imports "generic_os" and not "os" will be highly
> portable, and also highly constrained in what it can do.

Gee... sounds awfully useful; in fact, it sounds a lot more
useful than the current os module. But it's not very useful
if I have to make it up myself, since the purpose of such
a module is that the other programmers in the community
can recognize generic_os.listdir() as a well-known idiom.

>  > I hope the os module will be deprecated in future releases.
> 
> I don't think there is much chance of this happening, it it too widely
> used.

How about attempting to change that trend? i.e. how about
in stead of documenting os.getpgrp() with an availability
caveat, the os module documentation only lists the stuff
that's really expected to be portable, and has a note
at the bottom ala:

	the os module may contain other implementation-defined
	objects; for example, on posix platforms, the contents
	of the posix module are shadowed in the os module.
	[... list the other cases ...]

-- 
Dan Connolly, W3C
http://www.w3.org/People/Connolly/



From davidopp at megsinet.net  Wed Aug 25 05:51:11 1999
From: davidopp at megsinet.net (David Oppenheimer)
Date: Wed, 25 Aug 1999 05:51:11 -0400
Subject: Newbie Needs Direction
Message-ID: <37C3BC8F.1DFFC151@megsinet.net>

I have decided to learn python and Tcl.  Are there any great tutorials
out there for python?  I would also appreciate links to any resources
you may know about that are full of add-on tools, etc for python.  I am
also interested in wxPython.  I would like to also know what kinds of
programs people are designing with python.  And your comments on the
strengths and/or weaknesses you perceive in the language.   What
compilers are compatible, etc.?

Thanks,

David O.




From skip at mojam.com  Mon Aug 16 17:39:02 1999
From: skip at mojam.com (Skip Montanaro)
Date: Mon, 16 Aug 1999 16:39:02 -0500 (CDT)
Subject: Bug in struct module?
In-Reply-To: <37B8759C.D0DF1B55@bigfoot.com>
References: <37B8759C.D0DF1B55@bigfoot.com>
Message-ID: <14264.33395.160839.984120@dolphin.mojam.com>

    >>>> calcsize('hd')
    >> 16
    >>>> calcsize('dh')
    >> 10
    >> That seems like a bug to me.

Only if the native structs in C would be represented differently, and even
perhaps not then.  On my Linux box, the following short C program displays

    short size: 2
    double size: 8
    foo size: 12
    bar size: 12

when run.

    main() {
      struct {
	double f;
	short s;
      } foo;

      struct {
	short s;
	double f;
      } bar;

      printf("short size: %d\n", sizeof(short));
      printf("double size: %d\n", sizeof(double));
      printf("foo size: %d\n", sizeof(foo));
      printf("bar size: %d\n", sizeof(bar));
    }

It's possible that your compiler forces doubles onto eight-byte boundaries
but will truncate shorter fields at the end of structs.  On my machine it
appears to to pad everything to four-byte boundaries.  Compile and run the
program and see what you get.  Also, note that the struct module tries to
mimic the behavior of the C compiler.  From the struct module documentation.

    This module performs conversions between Python values and C structs
    represented as Python strings. ....

    By default, C numbers are represented in the machine's native format and
    byte order, and properly aligned by skipping pad bytes if necessary
    (according to the rules used by the C compiler).

That may well not be how VB does it, so you can't just assume that the
spacing the struct module gives you is how the VB object will be layed out.

Skip Montanaro	| http://www.mojam.com/
skip at mojam.com  | http://www.musi-cal.com/~skip/
847-971-7098





From oleg at pobox.com  Thu Aug 19 18:31:49 1999
From: oleg at pobox.com (oleg at pobox.com)
Date: Thu, 19 Aug 1999 22:31:49 GMT
Subject: 'if' as a _function_, in Scheme and Python
Message-ID: <7pi0kl$ko8$1@nnrp1.deja.com>

In most languages, 'if' has a distinguished status, of a statement or
a special syntax form. One wonders whether 'if' can be a regular
function. This question has a particular relevance to Python as it
lacks conditional expressions (as a ternary (?:) operator in C). IF
_can_ be a regular function, provided the evaluation of its 'branch'
arguments could somehow be delayed. Lazy languages like Haskell are
never in a hurry to evaluate arguments of a function call; in these
languages, IF indeed is no different than the other functions. In
eager, call-by-value languages the evaluation delay has to be spelled
out explicitly.

For example, in Scheme:
        (define (my-if condition then-branch else-branch)
           ((or (and condition then-branch) else-branch)))

on stipulation that the arguments of this function are wrapped in
lambda-expressions (i.e., thunks):

        (my-if (> 2 0) (lambda () (+ 2 3)) (lambda () (/ 1 0)))
        (my-if (pair? x) (lambda () (car x)) (lambda () #f))

BTW, in the my-if definition above, 'and', 'or' can be regular
functions rather than special forms. To show it clearer, let me
re-write my-if as

        (define (my-if condition then-branch else-branch)
           ((cdr (assq (eq? condition #f)
                (list (cons #t else-branch) (cons #f then-branch))))))

This definition is _purely_ functional: my-if is composed of regular
functions only; it does not include any special forms.

Surprisingly, the same technique _literally_ applies to Python. In
this language, the "if function" has more than an academic interest as
an 'if statement' in Python is not an expression (that is, does not
yield a value). And a ternary ?: operator is absent. However, the
following expression makes up for this:

x = eval((lambda: exp1, lambda: exp2)[not condition].func_code,vars())

this expression is equivalent to
        if condition: x = exp1
        else: x = exp2

No matter what object the 'condition' is or yields, "not condition" is
always 0 or 1. It has the value of 1 if and only if the 'condition'
evaluates to a "false object".  Like the Scheme my-if function, the
above Python expression is purely functional -- it does not include
'or' and similar shortcut operators. The 'if' expression appears so
complex because Python lacks truly lexical scope. Still, separate
local bindings can be imported into a nested function via default
arguments. The trick with eval shown above is a slightly more generic
version of emulating nested lexical or dynamic scopes. The eval
function should not impose a big overhead as it is given an already
compiled expression (code object). In this case, the eval acts as a
generalized function call.

Here is a more complex and real example:
   http.putrequest('POST',
     eval((lambda: "http://%s:%d%s" % (remote_host, remote_port,
server_url),
           lambda:  server_url)[not proxy_name].func_code,vars())

In this example, the conditional expression saves a temporary
variable, which will otherwise be required and has to be allocated and
assigned to. Conditional expressions help reduce pollution of the
namespace.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From evan at tokenexchange.com  Tue Aug 24 10:32:04 1999
From: evan at tokenexchange.com (Evan Simpson)
Date: Tue, 24 Aug 1999 09:32:04 -0500
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu> <yaZu3.23$2y4.21931@news-dal.corridex.com> <37C1B864.970C63DC@compaq.com>
Message-ID: <b2yw3.31$2y4.39902@news-dal.corridex.com>

Greg Ewing <greg.ewing at compaq.com> wrote in message
news:37C1B864.970C63DC at compaq.com...
> Evan Simpson wrote:
> >
> > A syntax for expressing explicit tuple
> > unpacking has been proposed (fn(*t)), and it remains to be seen whether
it
> > will enter the language.
>
> I'm working on an implementation of that, by the
> way, so it may be possible for people to try it
> out soon.

Sweet!  Are you doing dicts too?  I await the day when I can litter my code
with calls like "fn(1, *twople, 'with a melon?', *threeple, **websters,
**randomhouse)".

> > There are no compound
> > assignment operators since there are no assignment operators, period.
>
> I don't think it's helpful to jump on that particular
> terminological detail. The people who ask "why is there
> no += operator" aren't using the word "operator" in its
> precise technical Python sense.

Somebody else had already given the clear, helpful answer, and the repeated
references to Python statements as "functions" got me all pedantic.  I'm
better now.

> I'd say quite the opposite - this is a rare case in
> Python where IMplicitness loses! If the distinction
> between integer and floating point division were
> made explicit, by having different operators for
> them, this would cause much less grief.

You're only talking about division, right?  Now that I think about it,
Pascal's distinction between '/' (fp) and 'div' (int) division is a nice
one.

not-that-we-can-do-much-about-it-now-ly y'rs
Evan Simpson





From xah at best.com  Sun Aug 22 00:19:47 1999
From: xah at best.com (Xah)
Date: Sat, 21 Aug 1999 21:19:47 -0700
Subject: Why use Perl when we've got Python?!
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk>
	 <tRDs3.1443$J72.232020@news.itd.umich.edu>
	 <37B4AA81.DE8D6194@basho.fc.hp.com> <37b4adea@cs.colorado.edu>
Message-ID: <37bf7a63$0$8462@nntp1.ba.best.com>

----------
>From: Tom Christiansen <tchrist at mox.perl.com>
>Newsgroups: comp.lang.perl.misc
>Subject: Re: Why use Perl when we've got Python?!
>Date: 1999, Aug 13, , 5:44 PM

> > ...
> :
> :The Python FTP client module is designed to be a base class, not a user
> :class.
>
> Man, that's just superclever.  You really expect to stuff all this
> complete OO crudolo down some poor script kiddie's  throat complete
> just because they want a library that's simple to use?  That's nutty.
> If that's all Python has, it's no wonder why they use Perl.  Sheesh.
> Regular humans need to program, too, not merely the duly annointed
> OO indoctrinees.
>
> --tom
> --
>     "Don't wear rollerskates to a tug-of-war." --Larry Wall

Exactly. For the same reason I pity how far our society has advanced. I
remember we use to use Roman Numerals. When the Arab number system were
introduced, millions of people complained, including me. Similarly, when
typewriters are invented, I and others got extremely offended. I heard that
one guy said something like "no thank you, I can read hand-written letters
just fine". Then there's the electronic calculator shit. I am a hard-core
slide ruler user. I don't need sticking "calculator", and I don't want to
hear theoretical advantages. Why change when it ain't broken? I can go on
and on, but the most infuriating thing is the propagation of math in the
last few centuries. They fucking ram that shit down every one's throat,
including my children. I couldn't myself deal with infinity clearly.

I don't want to be tooo bitter, since it's all in the past. Let bygones be
bygones, and live today. I think the important thing for us programers today
is to vehemently support the Perl principles, and let no computer scientists
gave us away to some fancy theories. Let's try to stay 100% pure natural and
human. No this functional or OOP shit.

 Xah
 xah at best.com
 http://www.best.com/~xah/PageTwo_dir/more.html
 --
      "Don't shit in your pants" -- mom



From jstevens at basho.fc.hp.com  Fri Aug 13 23:07:44 1999
From: jstevens at basho.fc.hp.com (John W. Stevens)
Date: 13 Aug 1999 20:07:44 -0700
Subject: Why use Perl when we've got Python?!
In-Reply-To: <199908132020.OAA03738@jhereg.perl.com>; from "Tom Christiansen" at Aug 13, 99 2:20 pm
References: <37AAFAD1.A032765B@NOSPAM.strs.co.uk> <37B286D2.A71C3239@NOSPAM.strs.co.uk> <37b30253@cs.colorado.edu> <37B49CBA.772D9783@basho.fc.hp.com> <37b4a18f@cs.colorado.edu>
Message-ID: <37b4cf70@cs.colorado.edu>

> 
>      [courtesy cc of this posting mailed to cited author]
> 
> In comp.lang.perl.misc, 
>     "John W. Stevens" <jstevens at basho.fc.hp.com> writes:
> :What you seem to be complaining about is called "polymorphism"
> :in the OO world.
> 
> Perl is not about shoving an OO agenda down someone's throat.

I note, with surprise, the slant you put on my reply.

If I were to slant back, I would say that Python is not about
shoving some line-noise looking procedural agenda down somebodies
throat.

But, I won't.

Don't like OO?  Fine.  Don't use OO.

But if you do like OO, don't Perl.

John S.




From aa8vb at yahoo.com  Tue Aug 17 10:29:11 1999
From: aa8vb at yahoo.com (Randall Hopper)
Date: Tue, 17 Aug 1999 10:29:11 -0400
Subject: why are dlopen flags different between 1.5 and 1.5.2?
In-Reply-To: <m3oggkvu70.fsf@atrus.jesus.cam.ac.uk>; from Michael Hudson on Sat, Aug 07, 1999 at 10:03:31AM +0100
References: <"99Aug6.152924pdt.292365"@ngsun1.parc.xerox.com> <m3oggkvu70.fsf@atrus.jesus.cam.ac.uk>
Message-ID: <19990817102911.B2698559@vislab.epa.gov>

Michael Hudson:
 |Bill Janssen <janssen at parc.xerox.com> writes:
 |
 |> After a fair amount of grief, I found out that while Python 1.5
 |> specified RTLD_GLOBAL in its call to dlopen (in Python/importdl.c),
 |> Python 1.5.2 does not.  Is there any particular reason to change this?
 |
 |It was thrashed out here a while's back. www.deja.com is your
 |friend... I think there was also discussion on PSA-members or some sig
 |or other.
 |
 |I think the basic gist was that if not using RTLD_GLOBAL causes you
 |problems you can do something about it, whereas if using RTLD_GLOBAL
 |causes you problems you are screwed.

(A little late catching up on the Python folder)  

This bit me a few months ago as well.  If you're dlinking with C wrappers
which dlink with a library you don't have source for (which has undefined
unreferenced symbols that don't matter when linking with C programs),
you've got some work to do.

Either beg/borrow/steal the source and recompile it, define or remove the
undefined dynamic symbols one way or another, put the library in another
process, ...or just hack Python before you build it and remove that dlopen
flag.

Randall



From skaller at maxtal.com.au  Fri Aug 27 22:55:13 1999
From: skaller at maxtal.com.au (John Max Skaller)
Date: Sat, 28 Aug 1999 02:55:13 GMT
Subject: Why do we call python a scripting language?
References: <37c66347.518331@news.netset.com>
Message-ID: <37c748ad.1418655478@news.triode.net.au>

On Fri, 27 Aug 1999 10:18:55 GMT, guy_oliver at yahoo.com wrote:

>Why do we call python a scripting language?  
>
>For me, and for most of the people I try to tell about the benefits of
>python, the title 'scripting language' implies that its not a real
>language capable of real programs.  It implies to most of them that it
>is good for simple sysadmin and maybe short cgi scripts.

	Thats right, IMHO. Python (currently) lacks the facilities
required for system programming and programming in the large.

	First, it is not nearly fast enough. For example, my interscript
program is a literate programming tool written in python using
itself. Building interscript, using five or six passes, on my
120MHz Pentium running Linux takes several HOURS.
Using other literate programming tools written in C,
the build would be several SECONDS. Python is
hundreds of times too slow for interscript to be practical
for medium sized projects.

	Second, it lacks the data and control structures,
required for serious programming: they can be implemented,
but the implementations are clunky rather than slick.

	Third, there is very little static type checking,
which not only means python is error prone, and hard to
validate, compared to a language with a statically enforced
type system, but it also makes it hard to optimise

	There are related insecurities, such as the lack
of encapsulation. All these things make Python admirably
suitable for rapid prototyping and small programs,
but not nearly so well suited to building larger systems.

	I will hasten to add that quite a bit can be done
to overcome these problems while preserving many of the
nice features of python. For example, I'm working on a compiler
for Python which will do some static type checking by using
type inference. It is also possible to add new features
'in the spirit of python' which can enhance the facilities available
and make it more suitable for larger projects, both in terms
of efficiency and compile time error detection.

>I usually simply call it an interpreted object oriented programming
>language, as this does not carry with it the implication that its not
>a real language that cant do anything.  

	Python is a real language.

>I usually see people grouping languages into the compiled languages,
>such as c, c++, java, ada, etc, as being the only languages viable for
>large projects, important projects.  Python, perl, tcl, and others are
>grouped into the section of scripting languages for web sites, etc.
>This is not the case, is it?  

	Yes, usually, it is. There will be exceptions, where
performance is less of an issue, or where the requirements
are dynamic enough that the lack of compile time type
checking is outweighed by the need to check things at 
run time _anyhow_.

	Just look at an old version of Grail:
the performance is woeful, and it crashes so often
as to be unusable. Writing a 'modern' browser 
entirely in python (without a compiler) would be silly.
On the other hand, using Python to HOST 
the browser components, written in C, is eminently
sensible: top level architectures are well supported
by flexible dynamic systems like python, because it
makes reconfiguration, extension, etc, easy,
and fairly independent of the lower level, high
performance, components.

>Many of the programs I had tried to write before I
>knew python for my academic pursuits never got finished because I got
>tired of fighting with what ever language I was working with at the
>time.  

	Yes. And this is the status of interscript, because it
is getting too large for Python to handle. (quite apart from
the slow execution times).

	The moral is that you were fighting the language because
your application didn't suit the language.

>Since learning python, I have been able to write most of these
>programs that I never had the time to finish before in a single
>sitting.  Surely that makes python worthy of being thought of as a
>real, viable solution to just about any problem you choose?

	NO! you would be silly to try to implement an
operating system or compiler in Python, for example.
It makes Python a viable solution ... to those problems
for which Python is well suited. <grin>

>Are we doing Python and ourselves a disservice by calling python a
>scripting language?

	No. Python is a 'scripting' language, as opposed to
a systems programming language. It is good at small problems,
and it is good at gluing pieces of other stuff together quickly.

	Python is better (IMHO) at a larger scale of problem than
some other scripting language because:

	a) it has a nice syntax that is easy to work with

	b) it is built on a few simple concepts, so it is 
relatively easy to _reason_ about parts of a program
without having to hold all the code in your brain at once.
These concepts include things like a strong object basis,
the fact that assignments bind local variables only,
and the existence of packaging constructions like
classes, modules, packages, dictionaries, etc.

John Max Skaller                ph:61-2-96600850              
mailto:skaller at maxtal.com.au       10/1 Toxteth Rd 
http://www.maxtal.com.au/~skaller  Glebe 2037 NSW AUSTRALIA



From philh at vision25.demon.co.uk  Mon Aug 23 20:26:39 1999
From: philh at vision25.demon.co.uk (Phil Hunt)
Date: Tue, 24 Aug 99 00:26:39 GMT
Subject: *** Parrot-0.0.1 released ***
References: <935279776snz@vision25.demon.co.uk>
Message-ID: <935454399snz@vision25.demon.co.uk>

In article <935279776snz at vision25.demon.co.uk>
           philh at vision25.demon.co.uk "Phil Hunt" writes:
> It looks like version 0.0.1 of Parrot, my text-based GUI builder,
> will be ready within a day or so. It doesn't actually do anything
> useful yet; v0.0.1 is a technology demonstration version, which
> is intended to verify that the basic design is more-or-less
> sensible. Would anyone like to have a look at it, and possibly 
> suggest some improvements? If so, tell me, and I'll put it on
> my website.

Parrot is now on my website, at 
http://www.vision25.demon.co.uk/prog/parrot.html

-- 
Phil Hunt....philh at vision25.demon.co.uk




From donn at u.washington.edu  Tue Aug 10 15:31:10 1999
From: donn at u.washington.edu (Donn Cave)
Date: 10 Aug 1999 19:31:10 GMT
Subject: posix.error is a tuple, not an object or a string as docs say
References: <yttzozzbm11.fsf@gilgamesh.cse.ucsc.edu>
Message-ID: <7opulu$15fs$1@nntp6.u.washington.edu>

Ben Gertzfield <che at debian.org> writes:
| Today I was trying to use the posix.fdopen() function (through
| os.fdopen() as recommended in the documentation) so that I could open
| a pre-existing file descriptor passed in through sys.argv to a normal
| Python file object.
|
| Obviously I want to be able to catch errors in case the file
| descriptor passed through sys.argv isn't legal, so I read the nice
| documentation for posix.error:
|
|    error
|           This exception is raised when a POSIX function returns a
|           POSIX-related error (e.g., not for illegal argument types). The
|           accompanying value is a pair containing the numeric error code
|           from errno and the corresponding string, as would be printed by
|           the C function perror(). See the module errno , which contains
|           names for the error codes defined by the underlying operating
|           system.
|
|           When exceptions are classes, this exception carries two
|           attributes, errno and strerror. The first holds the value of
|           the C errno variable, and the latter holds the corresponding
|           error message from strerror().
|           
|           When exceptions are strings, the string for the exception is
|           'os.error'; this reflects the more portable access to the
|           exception through the os module.
|
| This doesn't seem to be quite true, however:
|
| Python 1.5.1 (#1, Dec 17 1998, 20:58:15)  [GCC 2.7.2.3] on linux2
| Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
| >>> import os
| >>> try:
| ...     input = os.fdopen(42)
| ... except os.error, error:
| ...     print error 
| ... 
| (9, 'Bad file descriptor')
|
| 'error', in this case, is a tuple, not an object. Am I incorrect in
| thinking this? If so, how do I access the aforementioned attributes
| errno and strerror?

Yes, I believe you are incorrect there.  'error' is indeed a class
instance, and it fooled you with its __str__ method.

   (9, 'Bad file number')
   >>> error
   <os.error instance at 14005e910>
   >>> error.__str__()
   "(9, 'Bad file number')"
   >>> print error
   (9, 'Bad file number')

But you're right, it has no "errno" attribute.  I guess this is a version
mismatch with your documentation.  If you upgrade to 1.5.2, it will have
those attributes - "errno", "strerror" and "filename".  1.5.1's os.error
still has only "args".  errno and strerror are args[0] and args[1].

	Donn Cave, University Computing Services, University of Washington
	donn at u.washington.edu



From bowman at montana.com  Fri Aug 20 00:17:19 1999
From: bowman at montana.com (bowman)
Date: Thu, 19 Aug 1999 22:17:19 -0600
Subject: Tk's default Toplevel - counterintuitive
References: <37BC6F2D.30A7@cs.bham.ac.uk> <37BC8D05.142E1DB0@aw.sgi.com> <37BC9CE8.67097FC6@vision.arc.nasa.gov>
Message-ID: <37BCD6CF.B7A186BC@montana.com>

Chad Netzer wrote:
> 
> PS.  What is the way to register a callback so that you can intercept the
> window manager's request to destroy a window?  Then you could just bind
> the sys.exit call to this on the Tk object...

def main():
	global top
	top = Tk()
	top.withdraw()
	
	mb = buildMenubar()
	mb.entryconfigure(2, activebackground='red')
	myTop = Toplevel(menu=mb)
	myTop.bind('<Destroy>', dying)

	
def dying(event) :
	print 'Toplevel is getting destroyed'
	top.quit()


The above snippet is a little ugly, but it works. Otherwise, if you kill
the visible window with the system box, the top process keeps on
chugging along. I like the Toplevel's menu setup rather than using
Menubutton's and the deprecated (atleast in the Tcl/Tk world)
tk_menuBar.




-- 
Bear Technology  Making Montana safe for Grizzlies

http://people.montana.com/~bowman/



From kaz at ashi.FootPrints.net  Thu Aug 26 13:38:32 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Thu, 26 Aug 1999 17:38:32 GMT
Subject: A Plea For A Return To Sanity!!!
References: <37bc1339@cs.colorado.edu> 	<37bc313b@cs.colorado.edu> 	<14268.17944.323020.242210@bitdiddle.cnri.reston.va.us> 	<019301beeaea$004c4d80$f29b12c2@secret.pythonware.com> 	<37bd4876@cs.colorado.edu> 	<MPG.1229cd5eed750a81989687@news.mop.no> <37C55649.8320449D@megsinet.net> <37C5659C.8892E52@callware.com>
Message-ID: <slrn7sautm.b88.kaz@ashi.FootPrints.net>

On Thu, 26 Aug 1999 16:04:44 GMT, Ivan Van Laningham <ivanlan at callware.com>
wrote:
>4)  I use Netscape mail.  Can someone out there tell me how to killfile
>Tom Christianson and Xah?  Thank you, and I apologize for the off-topic
>nature of this request.  Or is it?

The answer is, don't use Netscape as a newsreader. It's a piece of shit.
Get a real newsreader like slrn, trn, the GNUS thing for emacs, or
what have you.

I use slrn, which lets you killfile by subject, author or arbitrary pattern
matches on header and body. Plus it lets you scope the kills to be global or to
particular newsgroups. The kill isn't binary; you can assign an arbitrary
negative score, with -9999 meaning you never see the annoying idiot again. And
you can put in a date restriction, so that the ban will be automatically lifted
by a certain date. In the case of utter losers like Xah, it's ``-9999, all
newsgroups, and forever.'' ;)



From 75if at dsdfs.com  Sun Aug  1 17:20:46 1999
From: 75if at dsdfs.com (75if at dsdfs.com)
Date: 01 Aug 1999 14:20:46 PDT
Subject: over 18 only 31229
Message-ID: <01089914.2419@dsdfs.com>

Adults Only:

http://207.240.225.250/

*You must be 18 to click the link above.


\4<SN0Gs



From tim_one at email.msn.com  Tue Aug 17 00:32:50 1999
From: tim_one at email.msn.com (Tim Peters)
Date: Tue, 17 Aug 1999 00:32:50 -0400
Subject: Uglliness scale: Python <<< awk << p*** < SNOBOL4
In-Reply-To: <19990816205639474.AAA315.413@max41127.izone.net.au>
Message-ID: <000101bee869$9058c760$6e2d2399@tim>

[Robin Becker]
> I once described perl as being like awk on steroids and lsd

[John Machin]
> Actually perl is more like SNOBOL4 on sedatives.

While I'm tickled by the image, nothing is like SNOBOL4 -- not even its
modern successor, Icon.  SNOBOL4 was unique, and is still the Best Language
on Earth for writing string pattern-matching algorithms.

> [a nostalgic look at S4's pioneering features]
> ...
> It even had the ability to assign the result of a partial pattern
> match to a variable ... about 30 years later, Python got this in
> the re module's ?P<name> feature, and AFAIK perl doesn't have this
> even yet.

Neither does Python <wink>:  S4's "immediate assignments" allowed capturing
partial results even if the overall match failed.  Perl may yet get the
effect of that by allowing embedded Perl code in regexps, but Python likely
won't.

> However SNOBOL4 also had the ugliest syntax that I've ever seen in a
> language that was genuinely intended for serious work -- Intercal is
> far uglier but is a parody.

Its syntax was fine!  I expect that what you really object to is the absence
of control structures other than goto, and the LT/GE/etc spelling of
comparison operators.  That was common enough in its day, and even by the
time Pascal came around the keypunch I used still didn't have a semicolon
key.  It looks ugly in retrospect only because it is <wink>.

> For example [1]
> 	FLIP = LEN(*I) . HEAD LEN(1) $ X LEN(1) $ Y *LGT(X,Y)

All obvious to the most casual observer <snort>.  It's interesting that you
*still* can't spell this algorithm fragment with comparable ease in Python
or Perl!  Roughly, FLIP is a pattern fragment that skips the first I
characters of a string (or fails if the string is shorter than I chars),
assigns the next two characters to variables X and Y (or fails if there
aren't two more), and then succeeds if X is lexically greater than Y else
fails.  And the first I characters are assigned to HEAD iff the pattern as a
whole succeeds.  In context, the next line is

     STR FLIP = HEAD Y X

which matches FLIP against string STR, and interchanges X with Y if they're
lexically out of order (replacing the HEAD characters with themselves
unchanged).  In Python you'd be provoked to

    def flip(str, i):
        try:
            if str[i] > str[i+1]:
                str = str[:i] + str[i+1] + str[i] + str[i+2:]
        except IndexError:
            pass
        return str

To modern eyes I'm sure that looks clearer, but six delicate indexing
expressions are a half dozen chances to blow it that the S4 fragment can't
suffer.  I will admit that the call-by-reference Perl

sub flip {
    my ($s, $i) = @_;
    substr($$s, $i, 2) = $2 . $1 if $$s =~ /^.{$i}(.)(.)/ && $1 gt $2;
}

is much closer to the S4 <wink>.

> ...
> Did-anyone-we-know-write-a-SNOBOL4-mode-for-the-IBM-026-keypunch?-ly
> yours,

yes-but-it-was-in-fortran2-and-kept-blowing-the-tubes-ly y'rs  - tim





From da at ski.org  Fri Aug  6 13:13:41 1999
From: da at ski.org (David Ascher)
Date: Fri, 6 Aug 1999 10:13:41 -0700 (Pacific Daylight Time)
Subject: Proper application of the buffer interface
In-Reply-To: <14251.4680.760818.979812@dolphin.mojam.com>
Message-ID: <Pine.WNT.4.04.9908061010320.213-100000@rigoletto.ski.org>

On Fri, 6 Aug 1999, Skip Montanaro wrote:

> Sorry to be a numbskull, but since there is apparently little or no
> documentation for the buffer interface yet, can someone give me a thumbnail
> sketch of why we have it and what it's used for?  I can guess that there are
> some situations where using string objects as raw buffers could lead to
> misinterpretation of their contents, especially since not everything is
> US-ASCII anymore.  Beyond that, I'm not sure why there would be a
> distinction between buffers and strings.

I know of three (there may be more):

1) read/write, not just read (I'm using them to pass data to win32 calls
   which modify them, w/o new copies, etc.).

2) segmented architecture (non-contiguous strips of memory)

3) manipulating data which doesn't start its life as a PyString -- e.g. an
   mmap'ed file can expose a buffer interface, an Image can expose a
   buffer interface, a NumPy array can expose a buffer interface -- one
   can manipulate the data in those as buffers without having to copy the
   data. 

--david ascher




From kaz at ashi.FootPrints.net  Wed Aug 25 09:47:37 1999
From: kaz at ashi.FootPrints.net (Kaz Kylheku)
Date: Wed, 25 Aug 1999 13:47:37 GMT
Subject: equivalent of getch
References: <37C3C97A.59BF653E@acay.com.au> <7q0on8$h3i@news.rhrz.uni-bonn.de>
Message-ID: <slrn7s7t0m.f0q.kaz@ashi.FootPrints.net>

On 25 Aug 1999 12:48:40 GMT, Markus Fleck <fleck at informatik.uni-bonn.de> wrote:
>Gurvijay Singh Bhatti <khalsa at acay.com.au> wrote:
>> could someone tell me what is equivalent of getch in jpython? I want the
>> program to wait for the user to hit enter before going forward.
>
>AFAIK, this cannot be done in Java. I would
>be surprised if it could be done in JPython.

Note that waiting for enter doesn't require character-at-a-time input.  There
is no getch() function in C, but if all you want is to wait for enter, you
would use, say, fgets().



From fdrake at cnri.reston.va.us  Mon Aug  2 09:29:59 1999
From: fdrake at cnri.reston.va.us (Fred L. Drake)
Date: Mon, 2 Aug 1999 13:29:59 GMT
Subject: Libraries needed, please help!!!
In-Reply-To: <7o3moo$dse@news.go.dlr.de>
References: <7ns6o5$jo2@news.go.dlr.de>
	<004e01beda98$d14f9390$f29b12c2@secret.pythonware.com>
	<7o3moo$dse@news.go.dlr.de>
Message-ID: <14245.40279.79936.219772@weyr.cnri.reston.va.us>

Sven Drescher writes:
 > But how can I get the other files (libModules.a, libPython.a, libParser.a,
 > libObjects.a) from libpython1.5.a?

  Use libpython1.5.a instead of the others; it replaces all four of
them to make embedding easier.


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From fdrake at acm.org  Wed Aug 11 08:56:45 1999
From: fdrake at acm.org (Fred L. Drake, Jr.)
Date: Wed, 11 Aug 1999 12:56:45 GMT
Subject: Getting at line numbers with parsermodule
In-Reply-To: <000001bee3a8$ed930a00$ca2d2399@tim>
References: <000001bee3a8$ed930a00$ca2d2399@tim>
Message-ID: <14257.29453.920113.140089@weyr.cnri.reston.va.us>

Tim Peters writes:
 > would-the-manual-lie?-ly y'rs  - tim

  Hee hee hee...


  -Fred

--
Fred L. Drake, Jr.	     <fdrake at acm.org>
Corporation for National Research Initiatives



From I.Clarke at strs.co.uk  Wed Aug 18 08:00:07 1999
From: I.Clarke at strs.co.uk (Ian Clarke)
Date: Wed, 18 Aug 1999 13:00:07 +0100
Subject: Class Description
References: <37ba0d28@news.interq.or.jp>
Message-ID: <37BAA047.275A1536@strs.co.uk>

> ...could a class re-write itself, if properly messed with?

Do you mean like adding and removing methods and fields?

Ian.



From skip at mojam.com  Fri Aug 27 23:29:50 1999
From: skip at mojam.com (Skip Montanaro)
Date: Fri, 27 Aug 1999 22:29:50 -0500 (CDT)
Subject: Python in Blender
In-Reply-To: <37C6F48C.4C465AFB@fnmail.com>
References: <37C6F48C.4C465AFB@fnmail.com>
Message-ID: <14279.22415.399945.550048@dolphin.mojam.com>

    Per> Someone checked out Python in Blender already?

No.  Do you have a URL?

Skip Montanaro | http://www.mojam.com/
skip at mojam.com | http://www.musi-cal.com/~skip/
847-971-7098   | Python: Programming the way Guido indented...



From xdsi at xddainese.unive.it  Fri Aug 20 09:41:40 1999
From: xdsi at xddainese.unive.it (Diego Dainese)
Date: 20 Aug 1999 15:41:40 +0200
Subject: How to sell Python to your boss?
References: <11A17AA2B9EAD111BCEA00A0C9B41793034AAD03@molach.origin.ea.com>
Message-ID: <87wvuqefhn.fsf@blanka2.blankanet.it>

In comp.lang.python on Thu, 19 Aug 1999 17:41:07 GMT "Stidolph, David"
<stidolph at origin.ea.com> wrote:
[...]
> of development.  I also really appreciate the exception handling system in
> Python.  I use it in comparison to NOT using the C++ exception system.

Sorry for the OT, but why is it better to not use the C++ exceptions?

Bye,

-- 
Diego   | To reply remove the `x' and swap the | Sorry for my
Dainese | words around the `@' in the address. | bad English!



From nascheme at ucalgary.ca  Sat Aug 21 01:41:47 1999
From: nascheme at ucalgary.ca (Neil Schemenauer)
Date: 21 Aug 1999 05:41:47 GMT
Subject: More random python observations from a perl programmer
References: <37bc1339@cs.colorado.edu>
Message-ID: <7ple6r$kf0@ds2.acs.ucalgary.ca>

As no one seems to have written a reply to all of Tom's points, I
will make an effort.  Some of this information has been already
presented in other posts.  Hopefully this will be an accurate
summary.

Tom Christiansen <tchrist at mox.perl.com> wrote:
>GOTCHA: (medium)
>    The whole mutability versus immutability thing causes many 
>    confusions.  For example:
>	t = ([1,2],3)
>	t[0][0] = "fred"	# legal
>	t[1]    = "fred"	# ILLEGAL
>    I don't understand why tuples weren't just lists that weren't
>    somehow marked "read-only" or "constant".   That way you
>    could do the same with dictionaries or any other data types.

IMHO, this is one of the most important ideas to get straight
about Python.  All variables in Python are references.  For
objects that are immutable the semantics are the same as if
variables hold values.  This is good as no one wants the number
one modified when they aren't looking. :)

You have to be careful when modifying mutable objects.  Sometimes
you need to make a copy.  The copy module can do this for you.
If you are copying a list the "list[:]" idiom works just fine.  I
find I never use the copy module.

It is not too hard to remember what is mutable and what is not.
Basicly everything is mutable except numbers, strings, and
tuples.  Tuples are immutable so they can be used as keys in
dictionaries.  Numbers and strings are immutable for performance
reasons.  Check the FAQ for more details.

>GOTCHA: (low)
>    You can't use "in" on dicts.  Instead, you must use 
>	d = { "fred":"wilma", "barney":"betty" }
>	if d.has_key("fred"):	# legal
>	if "fred" in d:	        # ILLEGAL
>    I don't understand why this was done.

In the Python philosophy, every expression should have an obvious
meaning.  "in" in this case does not.  Perl has a different
philosophy here.

>GOTCHA: (low)
>    I don't understand why we have so many C operators, such as "&" and
>    "|" and "~", but we don't have "&&" or "||" or "!"  We also have "<<"
>    and ">>".  It's bizarre that the relational and "?:" are missing.
>    Python uses so much from C to make people see what they expect to see,
>    but this is missing.  Why?

I guess Python is not trying to look as much like C as possible
but instead be as readable as possible.  It does, however, borrow
C syntax.  I believe that Modula 3 has had a large impact on the
design of Python.

>GOTCHA: (high)
>    Local variables are never declared.  They just *happen*.  If a
>    variable is assigned to, then it belongs to that "scope".  If it's
>    only looked at, it may be from the current global scope.

The rule for determining if something is local is quite simple.
If a variable is bound anywhere in a function then it is local
unless declared global by the "global" statement.  Referencing a
local variable before it was bound used to give a strange
exception.  This has been fixed in the CVS version of Python.

>GOTCHA: (medium)
>    If you have a function that expects two arguments, such as:
>	def fn(x,y):
>	    return x + y
>    and you have a tuple of the two elements:
>	t = (1,2)
>    You can't just call that function:
>	print fn(t)
>    Or you get a "TypeError: not enough arguments; expected 2, got 1"
>    error.  You have to use this apply() function to get around this
>    issue:
>	print apply(fn,t)
>	3

I think most people feel this is a feature.  Python does not
generally do things implicitly.  A different philosophy then Perl.

>GOTCHA: (medium) 
>    Assignment is a statement, not an operator.  That means that
>    people are constantly writing loops that aren't really testing
>    the exit condition, as in:
>	while 1:
>	    line = os.readline()
>	    if not line: break
>	    .....
>    Because they can't write:
>	while (line = os.readline()):	# ILLEGAL
>    I feel that this hides the real test condition in a place
>    it doesn't belong.

On the up side, because assignment is a statement you can bind
multiple names at once:

	x = y = z = 0.0

>GOTCHA: (low) 
>    There are no compound assignment operators, like +=.

Again the semantics of such statements is not immediately obvious
so Python avoids them.  They may be eventually added to the
language though.

>GOTCHA: (low)
>    There are no loop labels, and therefore "break" and "continue" are
>    only through the next level.  This encourages the proliferation of
>    spurious boolean condition variables.  It was annoying when Pascal
>    made you do the same thing.  There is no "goto", which is how C
>    works around it.  As with many things in Python, here you force the
>    user to be tangled up with exceptions just to do very simple things.
>    That's not as clear a solution.

I have to agree with you on this one.  Using exceptions in this
case is not a Python-like solution.  In my experience I haven't
wished for this too often.

>GOTCHA: (high)
>    I had hoped that having *only* reference semantics would spare the beginner
>    from having to keep in mind the difference between a reference and its
>    referent as we have in C or Perl.  Alas, it is not to be.
>	 x = [1,2,3]
>	 y = x
>	 x[1] = "fred"
>	 print y
>	[1,"fred",3]

As I mentioned earlier, everything is a reference.  You just have
to know what is mutable and what is not.  Lists, obviously from
your example, are.


>COOLNESS:
>    You can "foreach" across multiple items.
>	a = [ [1,2], [3,4], [5,6] ]
>	for (i,j) in a:
>	    print i+j
>      3
>      7
>      11

This is called tuple unpacking.  It is definitely cool.  In the
newer versions of Python unpacking works on other sequences as
well.  For example:

	x, y, z = (1, 2, 3)
	a, b, c = [1, 2, 3]

This is especially useful when returning multiple values from
functions.

>COOLNESS:
>    Namespaces seem to be *the* thing in this language.  You can always
>    get a listing of anything's names/attributes, even all the built-ins
>    (which you can't do in perl) or all the names in the local scope
>    (which you also can't do in perl).  

Also because everything is looked up dynamicly, by rebinding
variables you can do really cool (or stupid) things.

>DISSIMILARITY:
>    There are no private variables in a module or a class.
>    They can always be fully-qualified and accessed from
>    without.  Perl can do this with file-scoped lexicals,
>    but still can't do so with data members (modulo obscure
>    tricks such as using a non-hash for an object).

In the words of Guido, we're all consenting adults. To make a
variable private proceed its name by one or two underscores.  The
two underscore thing is a new addition to the language.  If
someone monkeys with your private parts they get what they
deserve. :)

>GOTCHA: (medium)
>    Perl's hex() function is the opposite of Python's.  Perl's hex
>    function converts 22 into 37.  Python's hex function converts 37 into
>    0x22.  Oh my!  Like hex, Python has oct() completely backwards from
>    Perl.  Python's oct(44) returns 054, but Perl's oct(44) return 36.

Hmm, I'm not sure what Perl is doing here.  Python seems more
sensible to me.  The string 44 in the above Python example is a
literal.  Written as it is the Python parser treats it as a base
10 number.  If it was written as 054 then it would be treated as
a base 8 number.  oct() and hex() are functions that return a
number as a string in a certain base.  Maybe you want
string.atoi():

	>>> string.atoi('44', 8)
	36

>GOTCHA: (low)
>    Often Python's error messages leave something to be desired.
>    I don't know whether 

I don't find them too bad.  I guess I don't use Perl much so I
don't know what I am missing.  At least they are better than
SEGV. :)

>GOTCHA: (medium)
>    All ranges are up to but *not including* that point.  So range(3)
>    is the list [0,1,2].  This is also true in slices, which is the
>    real gotcha part.  A slide t[2:5] does not include t[5] in it.

This is explained quite well in the tutorial.  It makes a lot of
sense once you get used to it.

>DISSIMILARITY:
>    Python doesn't convert between strings and numbers unless you tell
>    it to.  And if you don't tell it to, you get an exception.  Therefore
>    you have no idea what these 
>	x = y + z
>	x = y - z
>	x = y * z
>    are really doing.  The first would concat strings or lists, the last would
>    repeat them.  Personally, I don't care for this, because I always
>    wonder what subtracting one string or list from another would be.

IMHO, + should not have been overloaded to mean concat.
Something like ++ would have been better.

>GOTCHA: (high)
>    This is a big surprise: strings are not atomic (although they are
>    immutable).  They are instead sequences of characters.  This comes
>    up in strange places.  For example:	
>	>>> for c in ("weird", "bits"):
>	...      print c
>	... 
>	weird
>	bits
>	>>> for c in ("weird"):
>	...      print c
>	... 
>	w
>	e
>	i
>	r
>	d
>    The second case autosplit the characters!

I think you are a bit confused by tuples.  In the first example
you are looping over a tuple of length two.  In the second case
you are looping over a string.  A length one tuple must end with
a comma.  This is weird but necessary for the parser to tell the
difference between a tuple and a parenthesized expression.

for loops over a sequence.  A string is a sequence of length one
strings (ie. characters).  If it wasn't for the tuple confusion I
don't think you would have found this too surprising.

>  Here's another:
>	>>> print map(None, "stuff")
>	['s', 't', 'u', 'f', 'f']
>	>>> print map(None, "stuff", "here")
>	[('s', 'h'), ('t', 'e'), ('u', 'r'), ('f', 'e'), ('f', None)]
>    (Python's map None is like Perl's map $_.) 

This is something a little different.  The map function accepts
more than one argument.  None behaves as the identity function
when passed to map.  I believe this is called zip() in functional
languages.

>GOTCHA: (high)
>    Because everything is a reference, and there's no way to dereference
>    that reference, it turns out that there is no trivial way to copy
>    a list!  This fails:
>	x = [1,2]
>	y = x
>    Because you don't get a new list there, just a copy to an
>    old one.  Suggested work-arounds include
>	y = x[0:]
>	y = x[:]
>	y = x + []
>	y = x * 1
>    This forces people to think about references, again.    
>    So much for lists being first class citizens!  Compare 
>    this with Perl's
>	@x = (1,2);
>	@y = @x;
>    Or even with references:
>	$x = [1,2];
>	$y = [ @$x ]; 
>    or 
>	@y = @$x;

Again with the references. :)  Don't think references.
Everything is a reference.  Forget about that.  Think
mutable/immutable.  Lists are mutable.  If you are modifying them
you better be careful about sharing them.  x[:] is an easy way to
make a copy.  For other stuff use the copy module.  You will use
it rarely.

Does Perl really make a copy of a list every time you assign it
to another variable?  How about other things like instances?
Using references always like Python does seems more natural to
me.  I usually want a reference, not a copy.  Also, what exactly
is a copy?

>GOTCHA: (medium)
>    Slices in Python must be contiguous ranges of a sequence.
>    In Perl, there's no such restriction.  

Use a dictionary if you want this.

>GOTCHA: (medium)
>    You can't slice dictionaries at all in Python.  In Perl, it's
>    easy to slice a hash, and just as sensible.

The semantics of this are not immediately clear.

>GOTCHA: (high)
>    As we saw with lists, because everything is a reference, and there's
>    no way to dereference that reference, this means that again there
>    is also no built-in, intuitive way to copy a dictionary.

There is a copy method in Python 1.5.2.  Also there is the copy
module.  Personally, I have never felt the need for this.  Maybe
my programming style is different than yours.

>GOTCHA: (high)
>    Lists don't autoallocate the way dictionaries do.

This catches a lot of range errors for me.  I like it.

>GOTCHA: (medium)
>    There's no way to set up a permitted exports list.  The caller may have
>    anything they ask for.  

Use underscores (and see above).

>COOLNESS:
>    DBM files seem (?) to automatically know about nested datatypes.   

You probably are talking about the pickle module.  Yes, it is
cool.

>GOTCHA: (medium)
>    Importing a variable only gives you a local copy.  In Perl, it makes
>    two names for the same object.

References, all of them.  Copying does not happen.

>GOTCHA: (low)
>    Because you can't declare a local variable, you can't create variables
>    local to something other than a module (ie. file), class, or function.
>    There are no loop-local variables; they would persist through the end
>    of the function.  You can't have a scope you can't name (well, almost;
>    see lambdas)

Python doesn't have arbitrarily nested scopes.  It would be a
nice feature.  In practice I do not find it to be a problem.
Functions and methods should be short.

>GOTCHA: (high)
>    Scopes don't nest in Python, but they they do in Pascal, Perl, or C.
>    This is supposedly "simpler", but it's very suprising in many ways:
>	x = 1			# global
>	def fn1(): 
>	    x = 10		# implicit local
>	    def fn2():
>		print x		# whose?
>	    fn2()
>	fn1()
>    The answer is that it prints 1, because fn2 has no x in its local
>    scope, so gets the global.  The suggested work-around is
>	    def fn2(x=x):
>    using default parameters.

See above (also the part about determining if a variable is
local).

>GOTCHA: (medium)
>    You can't cut and paste these examples because of the issue
>    of white space significance. :-(

I don't find this a problem.  In Vim >> and << work fine for me.

>GOTCHA: (low)
>    List objects have built-in methods, like 
>	l.append(x)
>    But string objects don't.  You have to import
>    from the string module to get them, and then they're
>    functions, not methods.

This is being fixed.  See the FAQ for more details.

>GOTCHA: (low)
>    You have insert and append methods for lists, but only
>    a del function.  Why is it a function not a method, and
>    why isn't it spelled out?  Apparently the fashionable way to do this is
>    now
>	a[2:3] = []
>    Which of course, deletes only one element.  ARG.

del is a statement.  It has to be so it can remove variable
bindings.  It can also do what you want as in:

	del a[2]
	del a[1:10]

Also see the remove method.  Python is not as orthogonal as it
should be regarding methods on objects.  That is being repaired.

>GOTCHA: (high)
>    There doesn't seem to be away other than using low-level hand-rolling
>    of posix functions to supply things like os.popen and os.system
>    a list of shell-proof arguments.  Looks like it always goes through the
>    shell.  This has security ramifications.  

I think there is a module somewhere that does this.  It is not in
the standard distribution however.

>GOTCHA: (high)
>    Because you can't use readline() to get a number, people seem to enjoy
>    calling eval() just to get a string turned into a number:
>	import sys
>	str = sys.stdin.readline()
>	num = eval(x)
>    This is scary.  Even scarier is the propensity for calling input(),
>    which auto-eval()s its input to make sure it's the right "type".
>    (Funny that a soi-disant "typeless language" should be so danged
>    picky about this.)  That means you see people write:
>	num = input("Pick a number? ")
>    But the person can supply as an answer 0x23 or 2+9 or pow(2,4)
>    or os.system("rm -rf *").  I am stunned.

You want int() or float() instead of eval().  For older versions
of Python use the string.ato* functions.  Things are not as bad
as you make them out.

>GOTCHA: (low)
>    Regexes default to emacs style, not egrep style!  Gads!  This is
>    surprising to anyone except an emacs weenie.  Sigh.  And just *try*
>    to read the manpage on the differences based on passed in arguments
>    establishing the style.  There aren't any manpages!  Have a nice day.

Use the re module for Perl style regexes.

>GOTCHA: (low)
>    An out-of-bounds list reference raises an "IndexError: list index out
>    of range" exception, but not if you use a slice to get at it!
>	t = range(5)
>	print t[2:17]
>      [2, 3, 4]

This sounds funny coming from a Perl guy.  Anyhow, look at the
protocol for slicing: s[2:] actually means s[2:sys.maxint].

>COOLNESS:
>    Relationals stack:
>	x < y < z
>    means 
>	x < y and y < z

Assignments too.

>GOTCHA: (high)
>    Python's lambda's aren't really lambdas, because they are only 
>    expressions, not full functions, and because they cannot see
>    their enclosing scope.  

Guido feels that lambda was a mistake.  I half agree.  We must
live with it now (or at least until 2.0).

>GOTCHA: (medium)
>    This is a tuple of two elements
>	(1,2)
>    But this is a tuple of one element:
>	(1,)
>    Whereas this is merely the expression 1:
>	(1)
>    Yes, the trailing comma counts.

You can always use the trailing comma if you like consistency.

>GOTCHA: (low)
>    Normally, a print statement in python adds a newline.  If you
>    don't want one, you need a trailing comma!

Use sys.stdout.write instead.

>GOTCHA: (low)
>    And I'd jolly well like to know why I wasn't allowed to use
>	print "int of %f is %.0f" % (2.5) * 2
>    or if needed, 
>	print "int of %f is %.0f" % ( (2.5) * 2 )

You've been had by the "tuple comma" gotcha.

	print "int of %f is %.0f" % ((2.5,) * 2)

works just fine.  The extra parentheses are needed because % has
a higher precedent than *.

>GOTCHA: (medium)
>    Anything that python doesn't like, it raises an exception about.
>    There is no fail-soft.  Even non-exception issues raise exceptions.
>    It's pervasive.  K&P curse languages that force people who want
>    to open() a file to wrap everything in exception code, saying that
>    "failing to open a file is hardly exceptional".  

This is a big feature to me.  I am not perfect.  perl -w would
probably help me but Perl didn't have that flag when I started
with Python.

>GOTCHA: (low)
>    sort and reverse are in-place.  This leads to verbose code, such as:
>	old = [1,2,3]
>	new = old
>	new.reverse()
>    Likewise for sort.  

This also leads to efficiency if you don't want a copy.  BTW, your
code is incorrect.  You want:

	new = old[:] # make a copy
	new.reverse()

Not so bad.

>GOTCHA: (low)
>    You have to compiled the definition for a  function before you may
>    compile the call it.
...
>    ADDENDA: It turns out that def() happens *AT RUN TIME* in python.
>    That's the root of this problem.  In Perl, a sub{} is a compile-time
>    thing.

Python does a lot of stuff at runtime.  It is a source of power
and of slowness. :)

>GOTCHA: (low)
>    When you need to make an empty copy of the same type, you write
>	x = y[:0]
>    So much for being typeless. Sigh.

I don't follow you here.  All Python objects have a type.  Can
you give me an example of where you would use this code?


I seem to have missed a few of Tom's points.  Hopefully other
people have addressed these.  Regarding the flames your posts
seem to attract, sorry about that.  Your tone seems to set some
people off, especially those not accustomed to your style.  Try
to be less confrontational, especially when new to a group.
Everyone else, try harder to get along.


    Neil



From jpeterson at office.colt.net  Mon Aug 23 03:46:54 1999
From: jpeterson at office.colt.net (Jon Peterson)
Date: Mon, 23 Aug 1999 07:46:54 GMT
Subject: Why use Python when we've got Perl?
References: <37b4ba01@cs.colorado.edu> <7p49l5$uv7$1@nnrp1.deja.com> <37b634a2@cs.colorado.edu> <37b9a99b.635832158@nntp.ix.netcom.com> <37bfe42c$0$221@nntp1.ba.best.com> <7ppi33$8oe$1@msunews.cl.msu.edu>
Message-ID: <O%6w3.223$u07.1770@news.colt.net>

In comp.lang.perl.misc Shannon Watters <watterss at pilot.msu.edu> wrote:

> Xah <xah at best.com> wrote in message news:37bfe42c$0$221 at nntp1.ba.best.com...
> >
> > Perl wouldn't be a nice community without bad asses like Tom Christensan,
> > Chris Nador and some other regulars I don't want to mention now. We thrive

> It's easy to act like a tough guy over the net, isn't it?

Hell yeah - I mean, err, why else does anyone read c.l.p.m ?

Remind me to start a bot that autoemails people the contents of:
http://www.5sigma.com/perl/topten.html

preferably still in HTML format. Now that I think of it the worst part of the
new www.perl.com site is that it includes a pointer to c.l.p.m. without an
attached health warning. And, amusingly, a search for 'perl community' turns
up precisely no documents. Well well well.

[runs off to his friendly London Perl Mongers group for sanity and useful help]





From news at dorb.com  Fri Aug 13 19:05:54 1999
From: news at dorb.com (Darrell)
Date: Fri, 13 Aug 1999 19:05:54 -0400
Subject: Question about order in lists
References: <DH%s3.206$O02.12373@typhoon.nyroc.rr.com> <NDBBIAPHEKGOBIHCBLIIMEECCAAA.mcfletch@vrtelecom.com>
Message-ID: <wx1t3.410$8P.14670@typhoon.nyroc.rr.com>

You win.
Map is very fast in this case.
Although something intresting happend when I used
l1[:] to create l2
##################
import operator, time

def mapIt(l1, l2):
    map( operator.add, l1,l2)


def loopIt(l1, l2):
    x=[]
    append=x.append
    for i in range(len(l1)):
        append(l1[i]+l2[i])


sz=1000000
l1=[]
for i in xrange(sz):l1.append(i)

if 0:
    # Time1: 5.75 sec
    # Time2: 27.70 sec
    l2=[]
    for i in xrange(sz):l2.append(i)
else:
    # Odd when using these huge numbers
    # This way of creating l2 ran faster ??
    # Time1: 3.25 sec
    # Time2: 28.50 sec
    l2=l1[:]

t1=time.time()
mapIt(l1,l2)
print " Time1: %.2f sec"%(time.time()-t1)
t1=time.time()
loopIt(l1,l2)
print " Time2: %.2f sec"%(time.time()-t1)

####################
# Time1: 3.25 sec
# Time2: 28.50 sec



--Darrell





From paul at prescod.net  Fri Aug 27 17:31:17 1999
From: paul at prescod.net (Paul Prescod)
Date: Fri, 27 Aug 1999 21:31:17 GMT
Subject: Why do we call python a scripting language?
References: <19990827112918.12652.rocketmail@web129.yahoomail.com>
Message-ID: <37C703A5.C9B18851@prescod.net>

Guy Oliver wrote:
> 
> Even after explaining it to them, and telling them my success
> stories, the improvements in productivity, the portability, etc, it
> seems that python suffers from the same view that hurts other
> interpreted languages, specifically, that interpreted languages,
> aka scripting languages, are for throw-away scripts, nothing more.

The irony is that Python would be more acceptable to some people if it
came with these shell scripts:

pycc: python -c "import %1"

pyrun: python %1

Anyhow, "everybody knows" that Java is the language of the future and
not only is it usually (byte)interpreted it is actually part of the Java
religion that it MUST be byte-interpreted. Cygnus' compiled Java is
heresy!

Seriously though, it would be worth some effort to make sure that
JPython and Cygnus's compiler are compatible. It would be especially
cool if old Python modules could be somehow linked in through the
embedded C runtime.

 Paul Prescod



From rainer2207 at my-deja.com  Sun Aug 29 05:52:56 1999
From: rainer2207 at my-deja.com (rainer2207 at my-deja.com)
Date: Sun, 29 Aug 1999 09:52:56 GMT
Subject: Newbie elementary question
Message-ID: <7qavtm$mha$1@nnrp1.deja.com>

Can someone please explain what "self" and
"master" are?

eg.
At the moment I have

class Interface:
      def __init__(self, master):
            ....
            ....

root = Tk()
inter = Interface(root)
root.mainloop()

I'm assuming that master is the root window and
self is the instance name of the window in the
class. But I've also seen master=None. Can
someone please explain this?



Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.



From mwh21 at cam.ac.uk  Thu Aug 26 06:29:28 1999
From: mwh21 at cam.ac.uk (Michael Hudson)
Date: 26 Aug 1999 11:29:28 +0100
Subject: emacs with python mode
References: <37C50DAD.840B67C5@xsoft.co.at>
Message-ID: <m3wvuirg1j.fsf@atrus.jesus.cam.ac.uk>

"Gerhard W. Gruber" <g.gruber at xsoft.co.at> writes:

[snip font coloring problems]

Have you tried global-font-lock? (Mind you I use GNU emacs on linux).

> How can I force emacs to expand all
> leading blanks to the according number of tabs or vice versa. I don't
> care wether I have tabs or blanks but I need to have the proper
> indention,

C-x h (select all)
M-x untabify (replaces tabs with an appropriate number of spaces)

> as you probably know this is a required part of the python
> syntax.

Umm, yes.

HTH,
Michael



From stephan at pcrm.win.tue.nl  Thu Aug 19 04:04:51 1999
From: stephan at pcrm.win.tue.nl (Stephan Houben)
Date: 19 Aug 1999 10:04:51 +0200
Subject: PIGgies?
References: <Pine.LNX.4.10.9908190118060.28527-100000@starship.python.net>
Message-ID: <m3d7wk19i4.fsf@pcrm.win.tue.nl>

Moshe Zadka <moshez at server.python.net> writes:

> On Thu, 19 Aug 1999, Kaz Kylheku wrote:
> 
> > On Wed, 18 Aug 1999 17:15:23 -0700, John Landahl <jpl at digex.net.nospamplease>
> > wrote:
> > >These are my opinions, of course, and I graciously solicit the opinions of
> > >others.  What do you think of the term "PIGgies"?  Do you feel that there is
> > 
> > Little piglets are cute animals. Besides, you obviously haven't seen an angry
> > boar charging at you at 100 miles per hour.
> 
> <philosophy>
> If a boar and a penguing would charge at each other,
> would anyone hear?
> </philosophy>
> 
> well-now-we-have-a-real-python-mascot-ly y'rs, Z.

Sure: a Python that has just eaten a pig.
Of course, grown-ups and Perl users would mistake it for a hat... [1]

Greetings,

Stephan

[1] Antoine de Saint-Exupery, "Le Petit Prince"